Anterior Shoulder Instability

Epidemiology, Mechanism of Injury, History, Physical Examination, and Imaging for Anterior Instability

LCDR James R. Bailey, MD
Barrett A. Little, MD
Kevin M. Dale, MD
Dean C. Taylor, MD


Anterior shoulder instability can be a difficult injury for both athletes to sustain and surgeons to treat. This chapter describes the epidemiology, mechanism of injury, history, physical examination, and imaging of this diagnosis.

Keywords: anterior shoulder instability; epidemiology; history; imaging; laxity; mechanism of injury; physical examination.


The views expressed in this article are those of the author(s) and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, or the United States Government. Dr. Bailey is a military service member. This work was prepared as part of his official duties. Title 17, USC, §105 provides that ‘Copyright protection under this title is not available for any work of the U.S. Government.’ Title 17, USC, §101 defines a U.S. Government work as a work prepared by a military service member or employee of the U.S. Government as part of that person’s official duties.

Imaging Introduction

  • Anterior shoulder instability is most common in young active males.

  • The position of the arm during a first-time dislocation is likely variable.

  • The history should include the patient age, activity level, mechanism of injury, number of dislocations or subluxations, and symptoms. The history is the most important information to diagnose this condition and will guide treatment.

  • Instability is a clinical diagnosis consisting of the symptom complex of pain, apprehension, the actual sensation or history of instability events, and reduced performance level.

  • It is important to evaluate for multidirectional laxity and to recognize multidirectional instability to maximize treatment. It should be recognized that hyperlax patients and those with multidirectional laxity can sustain an injury and develop unidirectional instability.

  • Physical examination should always start with the cervical spine and then move to the shoulder girdle. A systematic examination will ensure precise diagnoses.

  • Although much can be assessed with the appropriate plain radiographs of the shoulder, advanced imaging to evaluate for glenoid and humeral head bone defects may be required to fully define the pathoanatomy contributing to the instability.


The diverse spectrum of presentation contributes to the complexity of glenohumeral instability and leads to difficulties in diagnosis and management. It is important to distinguish between instability and laxity of the shoulder. Instability is characterized by the presence of symptoms in conjunction with abnormal laxity ( ). Failure of static and dynamic glenohumeral stabilizers leads to instability, which can be unidirectional or multidirectional with the most common pattern being anterior unidirectional instability. Anterior unidirectional instability can be the result of repetitive microtrauma or a single macrotrauma as in an acute anterior shoulder dislocation.

The incidence of shoulder dislocations in the United States is 23.9 per 100,000 person-years with the overwhelming majority (89%–98%) being anterior shoulder dislocations ( ). This is consistent with data obtained by on traumatic shoulder dislocations in Sweden. Put in perspective among other common orthopedic conditions presenting to emergency departments (EDs) in the United States, the incidence of shoulder dislocation is far less common than distal radius fractures, 62 per 100,000 person-years, and ankle sprains, 215 per 100,000 person-years ( ). However, the true incidence of anterior shoulder instability may be higher because of shoulders that spontaneously reduce and may never present to an ED. In a military epidemiologic study reviewing all U.S. Military Academy cadets who sustained a shoulder injury during one academic year, 85% of their instability events were subluxation with spontaneous reduction vs 15% dislocation ( ).

Age is an important factor in shoulder instability because the mechanism of injury, pathoanatomy, and recurrence rate differ among age groups. Almost half of dislocations occur in individuals between the ages of 15 and 29 years ( ). Described by Bankart as the “essential lesion” of shoulder instability, the vast majority of this young population who sustain an anterior shoulder dislocation will have associated anterior capsulolabral tears ( ). Young age at the time of initial injury is the most consistent and significant factor for risk of recurrence. The rate of recurrent instability ranges from 48% to 96% in this age group ( ). Approximately 20% of all shoulder dislocations occur in individuals older than 60 years of age ( ). In those older than the age of 40 years, 35% to 86% will have an associated rotator cuff tear ( ). This age group has a much lower recurrence rate than younger individuals, with rates as low as 4% ( ).

Gender and activity level are two other epidemiologic conditions that influence the incidence of anterior shoulder instability. Overall, males account for more than 70% of dislocations and are 2.5 times more likely than females to sustain a shoulder dislocation ( ). In collegiate athletes at one institution, the highest incidence of shoulder instability occurred in football players and wrestlers ( ). The majority of anterior shoulder instability occurs in active males, especially those who participate in contact sports.

Mechanism of Injury

The mechanism of injury in traumatic anterior shoulder instability can be repetitive microtrauma or, more commonly, a discrete traumatic event, leading to recurrent instability. In traumatic anterior unidirectional instability, the widely accepted position of the shoulder during a dislocation is abduction, extension, and external rotation (ER). This position is the point at which the anterior band of the inferior glenohumeral ligament (IGHL) is under maximum tension and is the primary restraint to anterior humeral translation ( ). The classic Bankart lesion is an avulsion of the IGHL complex from the anteroinferior glenoid, which likely occurs in this position when the humeral head is levered out of the glenoid. evaluated the arm position of anterior shoulder dislocations under anesthesia in patients with recurrent dislocations and concluded that at 90 degrees of forward elevation in the scapular plane, anterior translation was maximal at 26 degrees of ER. Only one shoulder showed anterior translation at maximal ER. This study suggests that the position during initial injury is likely variable. Another factor that supports varying positions during subluxation or dislocation is anterior instability in boxers occurring when missing a punch as described by . The unexpected unopposed anterior momentum of the upper extremity leads to anterior subluxation or dislocation. Owens et al also reported the most common mechanisms of injury in the same population of students at the United States Military Academy to be falls (15%) and collisions (14.5%). The mechanism of complete anterior shoulder dislocation is most commonly a discrete traumatic event, but the position of the arm during injury may vary. A recent study by used three-dimensional (3D) computed tomography (CT) scans to evaluate the position of Hill-Sachs lesions produced from dislocations of the shoulder in various degrees of abduction. They found that dislocations with the shoulder in a position of abduction greater than 60 degrees resulted in higher Hill-Sachs angles, or an angle more parallel to the anterior glenoid in a position of abduction, ER, placing the shoulder at a higher risk of engagement and recurrent dislocation. Shoulders that dislocated at lower abduction angles resulted in lower Hill-Sachs angles with less chance of engagement.


For anterior shoulder instability, the history is important for diagnosis and guiding treatment of the patient. Patient demographics are often the first part of the history that the clinician will obtain. The mechanism of injury should be investigated as to whether it was a traumatic injury or a spontaneous event. Symptoms at the time of injury should also be explored. Finally, history of instability of other joints or connective tissue diseases in the patient or family members should not be overlooked.

Patient age is often the first demographic information obtained upon meeting the patient. Whether the injured upper extremity is the patient’s dominant arm is important. The history should not overlook if the patient sustained any other injuries at the time of the injury. If the anterior shoulder instability occurred during a sporting event, the patient’s preferred sport, the time of the season, and the patient’s goals for return to that sport or another sport should be investigated.

It is important to understand the mode of onset in evaluating a patient with shoulder instability. There are several key questions to ascertain:

  • 1.

    Was there a specific single traumatic event that led to the recurrent instability?

  • 2.

    Was it trivial? If so, this should make the examiner concerned that hyperlaxity is more the issue.

  • 3.

    Was a radiograph taken showing the shoulder dislocated?

  • 4.

    How long did it take for the patient to recover from the initial event? A complete traumatic dislocation will result in anteroinfero-capsulabral disruption and possible glenoid and humeral head injury and take several weeks to recover as opposed to a patient with multidirectional instability who can recover quickly.

  • 5.

    What was the arm position at the time of dislocation? In the presence of a Hill-Sachs lesion, this may be able to predict on-track versus off-track lesions ( ).

  • 6.

    What is the arm position that the patient fears will reproduce symptoms (e.g., Abduction/External Rotation (ABD/ER); Adduction/Internal Rotation (ADD/IR))?

  • 7.

    Does the patient have instability during sleep?

  • 8.

    Does the patient require reduction for the recurrent instability events, or do they reduce spontaneously?

  • 9.

    Has it become easier to provoke an instability episode? This can signify significant capsulolabral degeneration and bone loss.

Symptoms at the time of injury will help distinguish between an initial anterior instability event, recurrent anterior instability, and nonshoulder issues. Patients with a first-time anterior shoulder dislocation typically have severe shoulder pain until the humeral head is reduced into the glenoid. Also, they are unable to continue with the current sport or activity at the time of the injury. Patients with recurrent anterior instability will usually have less pain and may be able to reduce the dislocation themselves. It is possible for high-functioning athletes with recurrent anterior instability to return to sporting events after a recurrent instability event ( ). If neurologic symptoms of paresthesia or weakness after the shoulder has been reduced persist, then a workup of a neurologic injury is indicated. If a patient complains of swelling or symptoms provoked by temperature changes, then a vascular workup of the extremity is required.

For patients with recurrent anterior instability, it should be determined how the instability is affecting their activities of daily living along with their ability to compete athletically. Patients who continue with contact sports after an initial traumatic anterior instability event can go on to develop recurrent anterior instability after multiple dislocations and subluxations. On the other hand, patients should be evaluated for a family history of anterior instability or instability in multiple joints to determine connective tissue disorders (e.g., Ehlers-Danlos and Marfan syndromes). Patients with ligamentous laxity should not be ignored because they may develop acute anterior shoulder instability along with multidirectional laxity ( ).

Physical Evaluation

A good history is instrumental in determining the diagnosis, and a well-performed physical examination will confirm the diagnosis and identify other conditions of the shoulder and the neck. Although most shoulder examination findings are imperfect in regards to sensitivity and specificity ( ), a combination of thorough history, physical, and imaging findings will exponentially improve diagnostic accuracy. Because many shoulder examination findings lack absolute specificity, we prefer a systematic approach that involves observation, palpation, range of motion, strength testing, neurovascular examination, and provocative tests specific for the acromioclavicular joint, biceps, superior labrum, rotator cuff, and anterior, posterior, and inferior instability. Other tests can be included specific to the patient’s symptoms (e.g., suprascapular nerve compression, thoracic outlet syndrome). It is important to always start with the cervical spine and then move on to the shoulder girdle because radiculopathy can often manifest as shoulder pain.

Multiple tests for anteroinferior shoulder instability have been described in the literature. Various described tests work imperfectly on different patient populations (multidirectional laxity vs traumatic unidirectional, thin female gymnast vs large football linebacker, acute vs chronic), so it is important to have several provocative tests specific for shoulder instability. It is also important to remember that in the acute setting; it is usually not possible to perform all these portions of a complete physical examination. In these cases, the practitioner will have to rely on a good history and imaging to make the diagnosis. A good examination still allows one to rule out other conditions because it is always possible to have multiple conditions at once in the shoulder. Instability can be associated with SLAP (superior labrum anterior and posterior) tears, axillary nerve injuries, multidirectional laxity with unidirectional instability, and scapular dyskinesia.

Specific tests to elicit anteroinferior shoulder instability (also see ) include the following.

  • Video 1A.1

    Video representation of a shoulder examination focused on anterior and anteroinferior shoulder instability.

Anterior load-and-shift testing ( ): This tests passive translation of the humeral head on the glenoid. The test can be performed with the patient standing, seated, or supine. The authors recommend having the patient in a supine position with the scapula stabilized by the examination table and the injured shoulder just coming off the bed. The arm is flexed in the plane of the scapula, the elbow is bent at 90 degrees, and various degrees of abduction should be tested (0, 45, and 90 degrees). One of the examiner’s hands should be used to hold the elbow and apply a compressive load through the center of the humeral head into the glenoid as well as a posterior counterforce while the other hand is used to apply anterior (or posterior for posterior load-and-shift testing) force ( Fig. 1A.1 ). Another variation of this maneuver is to have the patient lie on the examination table in the lateral position with the noninjured shoulder down. You can then kneel on the bed with your leg and thigh stabilizing the scapula and both hands free to manipulate the shoulder. The degree of translation should be graded on a 3-point scale. Grade 1 translation up to, but not over, the glenoid rim. Grade 2 anterior load and shift indicates that the humeral head can translate over the glenoid rim but reduces spontaneously. Grade 3 is when the humeral head translates over the glenoid rim and does not spontaneously reduce. The authors recommend against the use of “+” or “–” when using this grading scheme. Again, a differentiation between laxity and instability should be made at this point and depends on pain ( ). As with all maneuvers, the other side should be examined to determine if this is a unilateral finding. It is important to note that the reproduction of symptoms is the most important part of load-and-shift testing in clinic. In the operating room under anesthesia, the amount of translation is most pertinent.

Fig. 1A.1

Anterior load-and-shift testing. The arm is flexed in the plane of the scapula, the elbow is bent at 90 degrees, and various degrees of abduction are tested (0, 45, and 90 degrees). One of the examiner’s hands should be used to hold the elbow and apply a compressive load through the center of the humeral head into the glenoid as well as a posterior counterforce while the other hand is used to apply an anterior force.

Inferior sulcus sign testing: The inferior sulcus sign test should be performed with the patient in a seated position and relaxed with the arms at the side in neutral rotation. After ensuring that the patient is not guarding, a downward traction force is applied to the distal humerus or through a bent elbow while watching the lateral edge of the acromion ( Fig. 1A.2 ). The amount of inferior sulcus is measured in centimeters from the inferior border of the acromion to the humeral head. Grade 1 is equivalent to 1cm; grade 2, 2 cm; and grade 3, 3 cm. Some believe that if this sulcus persists when performed in maximal ER that it is an indication of rotator interval and superior glenohumeral ligament incompetence; however, it has been our experience that most shoulders’ sulcus signs persist in ER. Also, it is critical to note if symptoms are reproduced by the maneuver. Patients with multidirectional instability, whose hallmark feature is inferior patholaxity, often complain of pain and instability when there is an inferior directed load to the arm (carrying heavy objects at the side, for example)

Fig. 1A.2

Inferior sulcus sign test. This test should be performed with the patient in seated position and relaxed with arms at the side in neutral rotation. A downward traction force is applied to the distal humerus or through a bent elbow (solid arrow) while watching the lateral edge of the acromion and looking for a dimple or sulcus formed between the humeral head and acromion (dotted arrow).

The Gagey hyperabduction test: First described by , this test is used to assess the integrity of the inferior glenohumeral ligament. This test is performed with the patient sitting comfortably with the arm in neutral rotation and at the side. The examiner stands behind the patient and with one hand stabilizes the upper shoulder while the other hand raises the elbow in pure abduction ( Fig. 1A.3 ). In their study, 85% of anesthetized patients with shoulder instability had passive abduction greater than 105 degrees.

Fig. 1A.3

The Gagey hyperabduction test. This test is performed with the patient sitting comfortably with the arm in neutral rotation and at the side. The examiner stands behind the patient and with one hand stabilizes the upper shoulder while the other hand raises the elbow in pure abduction. The degree of abduction is measured.

The John Feagin test: First described in book chapter in 1984 ( ), this maneuver has been often attributed to Dr. Feagin and can be particularly helpful when examining large, muscular patients. It is accomplished by placing the arm in internal rotation (IR) with elbow fully extended on examiner’s shoulder. After the patient is appropriately calm and relaxed, one of the examiner’s hands can hold the elbow in full extension while the other can translate the humeral head anteroinferiorly on the glenoid ( Fig. 1A.4 ). This assesses the amount of anteroinferior translation of the humeral head on the glenoid and can be graded using the same grading scale as anterior load and shift. It should also be noted if this maneuver replicates the patient’s symptoms. Similarly, inferior laxity can be tested with the abduction inferior stability (ABIS) test described by , which places the patient in the same position while the examiner provides an inferior-only directed force. They describe this test as a way to study the effect of scapular inclination on inferior glenohumeral stability.

Fig. 1A.4

The John Feagin test. The maneuver is accomplished by placing the arm in internal rotation (IR) with the elbow fully extended on the examiner’s shoulder. After the patient is appropriately calm and relaxed, one of the examiner’s hands holds the patient’s elbow in full extension while the other translates the humeral head anteroinferiorly on the glenoid. The amount of translation is measured, and it should be noted if this replicated the patient’s symptoms.

Apprehension test with or without Jobe’s relocation: This test is best performed with the patient in a supine position of the edge of the bed with the scapula supported and the shoulder just off the table ( ). The arm should be at 90 degrees of abduction and ER with the elbow flexed to 90 degrees (throwing position). The humerus is held by the examiner with one hand while the other hand gently externally rotates the shoulder through the arm. Having apprehension, or the “feeling” that the shoulder is going to subluxate or dislocate, is a positive finding ( Fig. 1A.5, A ). Applying an anterior force to the humerus can augment the apprehension test. Positive findings in midrange apprehension (in a lower degree of abduction and ER) is a concerning sign for possible bone loss. The Jobe relocation test ( ) is then performed in the same position. Instead of an anteriorly directed force, a posteriorly directed force is applied to the humerus. If the patient is having a sense of instability, this posteriorly directed force should make the shoulder feel stable again and resolve any related pain ( Fig. 1A.5, B ). Resolution of pain and apprehension is a positive finding.

Fig. 1A.5

A, The apprehension test. This test is best performed with the patient in a supine position of the edge of the bed with the scapula supported and the shoulder just off the table. The arm should be at 90 degrees of abduction and external rotation with the elbow flexed to 90 degrees (throwing position). The humerus is held by the examiner with one hand while the other hand gently externally rotates the shoulder through the arm. Pain or apprehension that the shoulder is going to subluxate or dislocate is a positive finding. Applying an anterior force to the humerus can augment the apprehension test. B, The Jobe relocation test. This is performed after the apprehension test. A posteriorly directed force is applied to the humerus to simulate reduction of the joint. If the patient is having a sense of instability with the apprehension test, this posteriorly directed force should make the shoulder feel stable and resolve any related pain.

Finally, if treatment requires surgical intervention, it is important to always perform an examination under anesthesia after the patient is appropriately anesthetized. This should be performed on both arms and include forward flexion, abduction, abduction-ER, abduction-IR, anterior and posterior load shift testing, and inferior sulcus testing. The amount of translation should be noted. After surgical stabilization, this can be rechecked gently to assess for resolution of laxity.


Plain radiographs should be obtained, at the time of injury and after reduction, to ensure the humeral head is actually reduced on the glenoid and to rule out any other bony abnormalities. The authors’ normal shoulder series includes anteroposterior (AP), Grashey (true AP), Stryker notch view, and axillary or West Point modified axillary view. The Stryker notch view specifically helps to identify Hill-Sachs lesions, and the axillary views ensure the humeral head is reduced in the glenoid. The West Point view was developed to highlight the area of the anteroinferior glenoid, compared with the anterior glenoid highlighted in the axillary view. One can appreciate avulsion fractures, calcification, and anterior bone loss in the chronic setting. In a cadaveric study by , good correlation of anteroinferior bone loss was noted between the West Point view and CT scans. If a large Hill-Sachs lesion is noted on plain radiographs, one should consider advanced imaging as larger Hill-Sachs lesions combined with glenoid bone loss can be a reason for recurrent instability after stabilization ( ). Indications for advanced imaging include revision, instability in sleep, seizure, midrange apprehension in physical examination, frequent instability events requiring manual reduction, and concerns for associated injuries. These indications are discussed further later.

Magnetic resonance imaging (MRI) is the standard for radiographic evaluation of the labrum and other soft tissue structures. An MR arthrogram can be helpful in visualizing the extent of labral tears, SLAP tears, and other possible soft tissue injuries such as a humeral avulsion of the glenohumeral ligament (HAGL) and anterior labroligamentous periosteal sleeve avulsion (ALPSA), which have both been associated with recurrent instability if not addressed ( ). ALPSA lesions are variants of the anteroinferior Bankart lesion where the labrum is displaced medially on the scapular neck. These should warn the surgeon to closely examine the amount of bone loss because they are associated with almost double the amount of bone loss compared with standard Bankart lesions ( ). It is important to note that there is some controversy on whether an MR arthrogram is needed for diagnosis of these lesions. In the acute setting, blood from the hemarthrosis is usually enough of a contrast medium to differentiate structures. It is the authors’ preference to obtain an MR arthrogram when possible. However, if advanced imaging is unavailable, a study in West Point cadets found a 97% incidence of complete anterior capsuloligamentous detachment in patients with a first-time anterior shoulder dislocation requiring manual reduction by a health care professional ( ). The capsuloligamentous lesions were observed during arthroscopic examination of these shoulders. echoed these numbers with 93% of their patients who sustained an anterior shoulder dislocation had evidence on MRI of a soft tissue or bony Bankart lesion. More recently, his group looked at 27 military members who sustained a first-time, traumatic, anterior subluxation and found a 96% rate of Bankart lesions on MRI ( ).

Computed tomography is a very helpful addition to preoperative workup, particularly in the setting of suspected bone loss, which is a known risk factor for recurrent instability ( ). 3D CT reconstructions are often very helpful in visualizing the extent of bone deficiency. Axial scans in the plane of the glenoid and three-dimensional rendered images help define the extent and orientation of glenoid bone loss as well as any concomitant humeral head bone loss or Hill-Sachs lesions ( ).

Lateral Decubitus or Beach-Chair Positioning

Matthew T. Provencher, MD
Emil Stefan Vutescu, MD
George Sanchez, BS


One of the key determinants for a successful arthroscopic anterior shoulder instability repair is proper patient positioning before surgery. To successfully prepare a patient for arthroscopic shoulder surgery, extensive familiarity with the two most popular preoperative setups, beach-chair and lateral decubitus positions, is absolutely necessary. This chapter reviews both setups extensively and offers tips and pearls for each. Following this, a comprehensive review of the Bankart repair completed in the lateral decubitus position is provided.

Keywords: arthroscopic shoulder surgery; Bankart repair; beach-chair position; lateral decubitus; shoulder instability.


  • Traumatic anterior shoulder dislocation may result in recurrent instability, ultimately leading to extensive soft tissue lesions, bone loss, and development of osteoarthritis ( ).

  • Several factors determine the outcome of an arthroscopic shoulder instability repair; one of the key factors is proper preoperative patient positioning.

  • For a successful treatment outcome, a thorough understanding of the features of and differences between the beach-chair and lateral decubitus setup is essential.

  • This chapter provides a comprehensive overview of both possible setups followed by a detailed description of the Bankart repair with key pearls when completed in the lateral decubitus position.

Lateral Decubitus versus Beach-Chair Position

Arthroscopic anterior shoulder stabilization can be performed in either the lateral decubitus or beach-chair position. Each option has several advantages and drawbacks.

Lateral Decubitus Position

The lateral decubitus setup is preferred by many surgeons for its superior visualization potential, especially of the inferior aspect of the glenohumeral joint and subacromial space. It also provides enhanced instrumentation access to all areas (360 degrees) of the glenoid labrum. Detractors of this position stress the increased effort with conversion to an open procedure and the need for a more complex anesthesia airway management.

Setup Description

After operating room bed setup ( Fig. 1B.1 ) and induction of general anesthesia, the patient is placed laterally on a standard surgical table. A large beanbag is the preferred option for support ( Figs. 1B.2 and 1B.3 ), but a pegboard can also be used. Straps and braces can be used around the torso to secure the patient to the table. The operative shoulder is exposed vertically. An axillary roll is used to protect the brachial plexus ( Fig. 1B.4 ). The patient is titled 20 to 30 degrees posteriorly to angle the glenoid parallel to the floor. A foam pad is used to keep the head in a neutral position, and the eyes are protected ( Fig. 1B.5 ). The nonsurgical arm is placed onto an arm holder or arm board ( Fig. 1B.6 ). The operative arm is positioned into a sleeve with the thumb directed upward and then connected to a traction system. The system uses pulleys with weights to adjust abduction and forward flexion of the shoulder. Ideal forward flexion is 20 to 30 degrees, and ideal abduction is 25 to 45 degrees ( Fig. 1B.7 ). Anterior and posterior drapes cover the entire shoulder, and a lateral strap is used to improve visualization of the glenohumeral joint ( Fig. 1B.8 ). An arthroscopy drape is used under the axilla to collect effluent. Incisions are done away from pustules or acne to avert Propionibacterium contamination.

Fig. 1B.1

Setup of the operating room with beanbag and sheet prior to lateral decubitus positioning.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.2

While vacuum suction is applied to the beanbag, the patient is held in place with the assistance of two individuals.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.3

To avoid pressure sores, the edge of the beanbag where the edge of the bag meets the legs should be appropriately padded.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.4

To avoid compression of the downside axilla, an axillary roll is used. This type is an inflatable axillary roll.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.5

To protect the eyes and maintain the head in neutral position, a foam pad (indicated with orange arrows ) is used.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.6

The nonsurgical arm is placed onto an arm holder or arm board. In this case, an arm board with foam pads and a strap (indicated with orange arrows ) is used to maintain the nonsurgical arm in place.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.7

When positioning the surgical arm into place, forward flexion should be between 20 and 30 degrees ( A ), and abduction should be between 25 and 45 degrees ( B ).

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.8

Final setup with the arm abducted about 50 degrees and a lateral strap for improved visualization of the glenohumeral joint.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Tips and Pearls

  • Use a draw sheet on top of the deflated beanbag and underneath the patient’s torso to enable maneuvering.

  • Place patient supine on the deflated beanbag prior to start of anesthesia induction.

  • Angle the nonoperative arm holder at 120 degrees to the table ( Fig. 1B.9 ).

    Fig. 1B.9

    Overview of the lateral decubitus position at final setup.

    From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

  • After the lateral decubitus position is obtained, suction the beanbag while the patient is held in place at the legs and trunk (see Fig. 1B.2 ).

  • Pad the legs appropriately to prevent common peroneal nerve injury or any leg compression injury ( Fig. 1B.10 ).

    Fig. 1B.10

    For prevention of leg compression injuries, a pillow under the bottom leg as well as a second pillow between the legs is used.

    From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

  • After the patient is secure, turn the operative table 90 degrees for easy access to both the anterior and posterior shoulder joint for final setup.

  • The patient’s shoulder joint should be held in traction through use of a balanced suspension.

  • The posterior portal is linear with the lateral edge of the acromion and should be drawn after traction is in place.


  • 1.

    Ease of execution of instability repairs and superior labral anterior posterior repairs

  • 2.

    Superior visualization of the glenohumeral joint and subacromial space

  • 3.

    Enhanced access to the anterior, posterior, and inferior areas of the shoulder joint

  • 4.

    Decreased risk of cerebral desaturation events

  • 5.

    Decreased surgeon fatigue because work is done near the waist level

  • 6.

    Collection of cautery bubbles out of the field of view


  • 1.

    Nonanatomic orientation.

  • 2.

    Conversion to an open procedure requires repositioning and may necessitate reprepping and redraping

  • 3.

    Difficult to assess range of motion directly after repair

  • 4.

    Potential increased risk of neurovascular injury caused by prolonged traction.


Intraoperative traction can cause excessive strain on the brachial plexus and may trigger transient nerve paresthesias ( ). Of greater concern, permanent brachial plexus injuries have been reported ( ). The current recommendation is to limit traction to 2 hours and use less than 7 lb of lateral traction and 12 lb of longitudinal traction.

Although the risk of thromboembolic events in shoulder arthroscopy is extremely low, there is an increased incidence of deep vein thrombosis and pulmonary embolism in the lateral decubitus position ( ).

Beach-Chair Position

Advocates of the beach-chair position for shoulder arthroscopy reference its ease of conversion to an open procedure as the main advantage. Other benefits include flexibility of the anesthesia management, anatomic shoulder orientation, and greater freedom of shoulder motion during the procedure. Complications associated with the beach-chair position include a potential increase in risk of cerebral desaturation events ( ).

Setup Description

After anesthesia induction, the patient is placed on a standard or “beach-chair” table in the supine position ( Fig. 1B.11 ). To get optimal visualization, the medial border of the scapular spine should be exposed. Therefore the patient is best situated close to the edge of the table or slightly off the table, depending on body habitus. Padding of all the bony prominences is then completed. The next portion of the technique is placing the table in 20 degrees reverse Trendelenburg, while the hip is flexed 45 to 60 degrees and knees are flexed to 30 degrees to ease the sciatic nerve through use of a leg positioner ( Figs. 1B.12 and 1B.13 ). A well-padded head-holding device protects the head while covering the eyes during surgery ( Fig. 1B.14 ). For stabilization of the scapula, one or two towels should be placed against the medial scapular border ( Fig. 1B.15 ). The nonsurgical upper extremity can be positioned onto an arm holder, an arm board, or a sling or tucked at the side. The operative arm can be left free (an assistant will be needed to apply traction) or held in place through use of a commercially available sterile arm positioning device controlled by an assistant or placed on a well-padded Mayo stand ( Fig. 1B.16 ). Drapes cover the entire shoulder both anteriorly and posteriorly. An arthroscopy drape can be used under the axilla to collect effluent. Incisions should be done away from pustules or acne to circumvent contamination with Propionibacterium spp.

Fig. 1B.11

A, With the assistance of two individuals, the patient is placed into the beach-chair position. B, One person lifts up the patient while the other secures the head.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.12

To ensure a comfortable position for the patient, a leg positioner is used ( A ) and will allow for the pelvis to be level and directly against the surgical table ( B ).

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.13

A, For final position, ensure that the leg pad is up against the buttocks and the buttocks and low back are not slumped. B, Ensure that no pressure is placed on the peroneal nerve at the knee.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.14

A and B, A foam facemask is used to secure the head. The neck should be positioned in a neutral position. C, The facemask should not cover any mucosal or sensitive areas.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.15

The back holder should have one to two towels placed against the medial scapular border for stabilization of the scapula during the procedure.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1B.16

Final positioning with the arm reduced to a pneumatic arm positioner.

From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Tips and Pearls

  • Ensure the arm positioner is situated on the table before onset of the case.

  • The table should be elevated to 60 to 80 degrees for shoulder arthroscopy.

  • Avoid having the patient too inferior on the table because this may result in pressure ulcers and neuropraxia.

  • A leg positioner can be used to maintain the pelvis level and against the back of the bed (see Fig. 1B.12 ).

  • Various available head-holding devices with straps and gel foam can be used to ensure neutral head and neck position ( Fig. 1B.17 ).

    Fig. 1B.17

    For securement of the head, a foam facemask may be used.

    From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

  • A wedge pillow can be used under the thigh (placed all the way up to the buttocks) to prevent pressure sores.

  • Place two towels posterior to the scapular spine against the table (see Fig. 1B.15 ).

  • Table should be flexed to confirm that the legs are level.

  • Pad the legs appropriately to prevent common peroneal nerve injury or any leg compression injury (see Fig. 1B.10 ).

  • Maintain even shoulder height to prevent brachial plexus stretching.

  • The operative arm should be in neutral position at the start of the arthroscopy and then manipulated as needed.

  • Use cerebral oximetry in high-risk patients to identify potential cerebral hypoperfusion during surgery.


  • 1.

    Ease of conversion to open procedure

  • 2.

    Upright, anatomic position of the glenohumeral joint—easier teaching and orientation

  • 3.

    Ease of recognition of external shoulder landmarks allows increased precision in portal placement.

  • 4.

    Potential decreased risk of nerve palsy ( )

  • 5.

    Flexibility of anesthesia, whether regional or general

  • 6.

    Easy airway access if complications arise during surgery

  • 7.

    Superior anterior shoulder portal access

  • 8.

    Increased range of motion of the operative arm


  • 1.

    Increased risk for cardiovascular complications such as cerebral ischemia

  • 2.

    Potential increased risk of air embolus

  • 3.

    Head acts as a mechanical block for posterior and superior portals; a transrotator can be used during posterior Bankart repair ( )

  • 4.

    Collection of cautery bubbles can cloud the subacromial space

  • 5.

    Challenging access to the inferior, superior, and posterior areas of the glenohumeral joint

  • 6.

    Increased cost because of the beach-chair attachment and mechanical arm holder.


Cerebral ischemia during deliberate hypotensive anesthesia is a potential risk in the upright beach-chair position. Cerebral desaturation events have a higher incidence in the beach-chair position versus the lateral decubitus position ( ). Of note, the anesthesia literature recommends maintaining systolic blood pressure at above 90 mm Hg and mean arterial pressure (MAP) less than 20% to prevent cerebrovascular events ( ).

Other reported complications include hypoglossal nerve palsy ( ) (attributed to repeated changes in neck position), greater auricular and lesser occipital palsy ( ) (caused by head holding device compression), and ventricular tachycardia ( ).

Senior Author’s Preferred Method: Bankart Repair Technique to Treat Recurrent Anterior Shoulder Instability in the Lateral Decubitus Position

An anterior shoulder dislocation even may result in a glenoid labrum avulsion known as a Bankart lesion ( Fig. 1B.18 ). The Bankart lesion causes an increase in anterior translation in all points of elevation, thereby generating recurrent instability ( ). It is now widely agreed that the proper treatment for this type of shoulder pathology is surgical repair, either open or arthroscopic.

Fig. 1B.18

A, Axial view of a magnetic resonance arthrogram demonstrating a soft-tissue Bankart tear anteriorly. The black arrow points to the anteroinferior labrum. B, Corresponding arthroscopic image of the anterior Bankart tear indicated by arrows . MGHL , Middle Glenohumeral Ligament.

From Provencher MT, Ghodadra N, Romeo AA. Arthroscopic management of anterior instability: pearls, pitfalls, and lessons learned. Orthop Clin North Am . Philadelphia: Saunders; 2010 ; 41(3): 325–37.

A recent meta-analysis suggests that the arthroscopic technique using modern suture anchor technique has similar outcomes to the open method ( ). Previous reports indicate that the open technique has a lower recurrence rate, especially in young male patients ( ). This is controversial, but when using proper indications and excluding patients with significant bone loss, the arthroscopic technique can provide an excellent outcome with less surgical morbidity. Arthroscopic repair of the Bankart lesion can be performed through either lateral decubitus or beach-chair positioning as previously described.

The senior author prefers the lateral decubitus position because of the superior visualization of the capsulolabral complex and lower instability recurrence rates when compared with the beach-chair setup ( ). Because of the greater exposure, particularly of the inferior aspects of the shoulder joint, the lateral decubitus position provides considerable instrumentation accessibility to the entire joint and labrum (360 degrees). In addition, this setup allows for lower suture anchor position on the glenoid, thus making the repair more easily secured.

Bankart Repair Technique

  • 1.

    Preoperative evaluation

    • Assessment of glenoid or humeral head bone loss

    • The most accurate method for measuring glenoid bone loss is a three-dimensional computed tomography scan with digital subtraction of the humeral head on sagittal oblique imaging. Magnetic resonance imaging or a magnetic resonance arthrogram may also be used.

  • 2.

    Evaluation under anesthesia

    • Confirms the direction of instability

    • Indicates the amount of translation in the anterior, posterior and inferior directions

    • Helps demonstrate the amount of capsular plication to be performed

  • 3.

    Patient positioning

    • Lateral decubitus position is preferred. Specific description of this setup is presented earlier in the chapter.

  • 4.

    Portal placement

    • First two portals—anterosuperior and posterior—to assess injury pattern. The glenohumeral joint is inspected for chondral damage. The posterolateral side of the humeral head is examined for potential Hill-Sachs lesions. Biceps tendon, glenohumeral ligaments, rotator cuff and anterior, inferior and superior aspects of labrum are also inspected for pathology.

    • The standard posterior portal in the lateral decubitus setup is in line with the lateral edge of the acromion and 1 cm inferior to the posterior tip. Placing this portal more lateral than the typical beach-chair portal allows for slightly downward trajectory on the glenoid surface and provides satisfactory access to the posterior glenoid rim for anchor placement.

    • The anterosuperior portal is placed right at the anterior lateral edge of the acromion. This will be the principal visualization portal of the procedure.

    • After Bankart lesion detection, an anterior midglenoid portal is placed in line with the 3 o’clock glenoid. An 18-gauge needle just superior to the subscapularis tendon can aid with placement of this portal ( Fig. 1B.19 ). The mid-glenoid portal with an 8-mm cannula acts as the working portal.

      Fig. 1B.19

      A, Outside view of shoulder demonstrating the arthroscope in the anterosuperior portal while the working portal is located at the anterior midglenoid portal and posterior portal. B, Corresponding arthroscopic view from anterosuperior portal demonstrating an anterior-inferior capsulolabral avulsion.

      From Provencher MT, Romeo AA. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

    • The last portal established is the posterolateral one. It is 4 cm inferior to the posterolateral border of the acromion, and its main role is to facilitate anchor placement in the inferior position of the shoulder.

    • Avoid the axillary nerve, which is 12 to 15 mm away from the articular margin of the glenoid, articular between the 5:30 and 6 o’clock positions.

  • 5.

    Glenoid preparation

    • The arthroscope is placed in the anterosuperior portal.

    • An elevator tool (through the midglenoid portal) is used to complete soft tissue elevation from the glenoid. Attention is placed on maintaining the integrity of the circumferential fibers of the labrum.

    • The labrum is peeled from the glenoid neck by using elevators and a small bone-cutting shaver.

    • Bleeding bed of bone for healing should be present.

    • Visualization of the posterior subscapularis muscle fibers indicates satisfactory glenoid preparation.

    • The labrum and capsule should be freely mobile before repair.

  • 6.

    Suture anchor placement and capsular plication

    • Anchor placement is always along the articular cartilage margin.

    • Anchors should be placed so that the sutures are perpendicular to the glenoid rim.

    • Standard repair consists of three or four anchors below the 3 o’clock position on the glenoid rim at a 45-degree angle to the glenoid surface ( Fig. 1B.20 ). Additional anchors are placed 5 to 7 mm apart (depending on the extent of the defect).

      Fig. 1B.20

      Arthroscopic view of an anatomic restoration of the labral-ligamentous complex following Bankart repair in the posterior viewing portal.

      From Provencher MT, Ghodadra N, Romeo AA. Arthroscopic management of anterior instability: pearls, pitfalls, and lessons learned. Orthop Clin North Am . Philadelphia: Saunders; 2010.

    • The first suture anchor is completed via the posterolateral portal.

    • The remaining suture anchors are knotless and done with labral tape through the midglenoid portal.

    • Capsulolabral tensioning is ideal in slight external rotation.

    • Before knot tying, sutures are pulled to confirm sufficient capsular tissue has been integrated (see Fig. 1B.20 ).

  • 7.

    Testing for stability

    • A gentle load-and-shift maneuver can be performed to be sure that patholaxity has been eliminated.

    • General range of motion is examined after portal closure and balance suspension removal. We check to be sure that there is at least 30 degrees of external rotation.

    • At the end of the procedure, the patient is placed in a padded abduction sling with a cold compression therapy device.

Arthroscopic Capsulolabral Repair and Reconstruction in the Beach-Chair Position

Peter D. Fabricant, MD, MPH
Frank A. Cordasco, MD, MS


This chapter reviews the indications, techniques, complications and post-operative rehabilitation of arthroscopic stabilization for anterior shoulder instability. Advanced arthroscopic techniques including double row labral repair and the addition of a posteroinferior labral repair and capsular plication are discussed for high-risk young athletes.

Keywords: anteroinferior capsulolabral repair; arthroscopic labral repair; Bankart repair; double row labral repair; labral repair.


  • Primary arthroscopic anterior stabilization may be routinely performed for recurrent anterior instability and uncomplicated revision surgery without significant bone loss.

  • Although the patient position in either the beach-chair (modified Fowler) or lateral decubitus position may be used to successfully treat glenohumeral instability arthroscopically, the beach-chair position offers some distinct advantages in some clinical scenarios. The use of an arm-holding traction device with a “bump” in the axilla facilitates access to the inferior aspect of the glenoid and capsule.

  • The beach-chair position is advantageous in that it allows for an easy transition between arthroscopic and open approaches without patient repositioning.

  • If a surgeon prefers the beach-chair position for rotator cuff repair, it may be advisable to use the same position for glenohumeral instability surgery, particularly in older patients in whom concomitant rotator cuff pathology frequently accompanies glenohumeral instability.

The use of arthroscopy for the treatment of shoulder instability has continued to increase as refined techniques have improved clinical outcomes. Arthroscopy provides several advantages in the treatment of glenohumeral instability, including improved visualization and clarification of pathoanatomy ( ), delineation and confirmation of preoperative imaging evaluation ( ), and minimally invasive treatment of pathology. Although open stabilization continues to have a role in the surgical management of specific subtypes of glenohumeral instability (e.g., significant glenoid bone loss or bipolar lesions, poor tissue quality, extreme hypermobility, revision surgery), advanced arthroscopic techniques continue to offer expanded indications for arthroscopic stabilization ( ). In addition to improved surgical visualization with minimal soft tissue dissection, there is a significant reduction in operative and hospitalization time and perioperative complications with arthroscopy compared with open surgery ( ). Although both beach-chair (modified Fowler) and lateral decubitus positioning have been described extensively, this chapter focuses on glenohumeral arthroscopy in the beach-chair position, as popularized at the Hospital for Special Surgery ( ). Open repair and reconstruction techniques are described in detail in the other chapters of this section. The epidemiology, pathoanatomy, patient evaluation, imaging, and surgical outcomes are also discussed elsewhere. The purpose of this chapter is to describe the indications and technique for arthroscopic capsulolabral repair in the beach-chair position.

Indications and Contraindications

The decision to use the beach-chair position for glenohumeral stabilization is based on the surgeon’s preference and has similar indications to arthroscopy in the lateral decubitus position for many cases. However, there are certain clinical scenarios when the beach-chair position is preferable. If during preoperative planning the surgeon identifies pathology that may require conversion to an open approach (e.g., significant glenoid bone loss or bipolar lesions, poor capsulolabral tissue, HAGL [humeral avulsion of the glenohumeral ligament] lesion), this may be easily performed without patient repositioning, but an open deltopectoral approach is not readily performed in the lateral decubitus position and would require patient repositioning intraoperatively. Additionally, although either position is commonly used for treatment of glenohumeral instability, treatment of rotator cuff tears is more commonly addressed in the beach-chair position. If a surgeon prefers the beach-chair position for rotator cuff repair, it may be advisable to use the same position for glenohumeral instability surgery, particularly in older patients in whom concomitant rotator cuff pathology frequently accompanies glenohumeral instability ( ).

With regard to indications for arthroscopic management of glenohumeral instability in general, many patients with glenohumeral instability expect arthroscopic stabilization. More important, therefore, is deciding which patients are not candidates for arthroscopic stabilization. Risk factors for failure include age younger than 20 years old, competitive overhead or contact athlete, shoulder capsular attenuation (caused by repetitive instability episodes, poor-quality capsulolabral tissue, or a genetic collagen disorder), significant bone loss on the glenoid or humeral sides, bipolar and engaging lesions, a Hill-Sachs lesion present on the anteroposterior (AP) radiograph in external rotation, and loss of the anteroinferior glenoid contour on the AP radiograph ( ). These factors have been used to develop the instability severity index score (ISIS), which identifies patients who may be treated more effectively with open stabilization ( ). When the interval of time between the dislocation and reduction approaches 5 hours or more, stabilizations that address bone loss, such as the Latarjet reconstruction, should be considered ( ). The number of formal reductions required after anterior dislocation has been shown to be directly correlated to failure after arthroscopic anterior stabilization and may warrant consideration for an open technique ( ). Recent literature indicates that more than one preoperative dislocation may increase the odds of failure of arthroscopic stabilization by 3.8 times ( ).

We believe that patients with demonstrable bone loss from the glenoid or humeral head warranting bone augmentation or transfer (e.g., Latarjet procedure [ ], Bristow procedure [ ], free allograft or autograft [ ]), patients with capsular injury, and patients requiring a revision stabilization who require an open approach should undergo diagnostic arthroscopy before open surgery to document all pathology ( ). In such cases or in the event the surgeon is unsure whether an arthroscopic or open approach is ideal, the beach-chair position is preferable because it allows for rapid conversion from arthroscopic to open surgery without patient repositioning if needed. This is important because arthroscopic examination may alter diagnoses in 40% of patients with atraumatic instability and in 8% of patients with traumatic anterior instability ( ). Although capsular avulsions of the glenohumeral ligaments have been successfully repaired arthroscopically, many arthroscopic shoulder surgeons still consider anterior HAGL lesions a relative contraindication to an all-arthroscopic stabilization. In the senior author’s practice, 80% of surgical cases of anterior instability are treated arthroscopically, 10% with open stabilization, and 10% with bone augmentation (majority Latarjet) reconstruction.

Surgical Technique of Capsulolabral Repair in the Beach-Chair Position

We perform primary arthroscopic anterior stabilization routinely for recurrent anterior instability and uncomplicated revision surgery without significant bone loss. We do not consider participation in contact or overhead athletics a contraindication to arthroscopic stabilization. We do counsel collision athletes that recurrence after stabilization depends on the repair type and on the activity level chosen after surgery. In general, we have been satisfied with our arthroscopic results (even in contact athletes) when the following conditions are met:

  • The patient has unidirectional instability of traumatic origin consisting of anterior or isolated anteroinferior instability without more than 2+ sulcus sign on examination under anesthesia.

  • Arthroscopic examination discloses a discrete Bankart lesion. The lesion in some cases has migrated and healed medially and inferiorly (ALPSA [anterior labroligamentous periosteal sleeve avulsion] lesion) ( ). When the capsule and labrum have healed in the medialized position, the capsule and labrum must be elevated off the glenoid and brought to a functionally appropriate position on the anterior-inferior glenoid rim.

  • There is no significant bone loss. Although historically, 20% to 25% glenoid bone loss was considered to be significant, reported that the critical level of bone loss may be lower and that in those with high level of mandatory activity, glenoid bone loss greater than 13.5% led to inferior clinical outcomes even without recurrent instability.

  • The capsule is robust, and the Inferior Glenohumeral Ligament (IGHL) complex does not exhibit signs of attrition.

  • No anterior HAGL is present.

  • No off-track or engaging bony Bankart, Hill-Sachs, or combined lesion is contributing to instability within a functional range of motion. have recently shown that glenoid bone loss of as little as 2 to 4 mm (8%–15%) in the setting of a Hill-Sachs lesion could compromise isolated Bankart repair and may require concomitant additional procedures.

It is important to be prepared to convert to open surgery if examination under anesthesia and diagnostic arthroscopy reveal any of the aforementioned diagnoses not amenable to successful arthroscopic reconstruction. As mentioned earlier, the beach-chair position allows for an easy transition between approaches without patient repositioning.

Regional anesthesia with an interscalene block combined with general anesthesia is administered. The patient is placed in the beach-chair (modified Fowler) position. Arthroscopic landmarks are palpated and marked on the skin of the shoulder, including the acromion, acromioclavicular joint, clavicle, and coracoid.

An examination under anesthesia is performed to confirm the pattern of instability because the examination under anesthesia is more sensitive in determining both the degree and direction of instability. The load-and-shift test is conducted, and translation is noted in the anterior, inferior, and posterior directions. Grading reflects the degree of humeral head translation anterior and posterior to the glenoid rim. The grade is 1+ if the translation of the humeral head is to the edge of the glenoid, 2+ if the humeral head can be subluxated over the glenoid rim but reduces spontaneously, and 3+ if a frank dislocation of the humeral head over the glenoid rim does not reduce spontaneously.

The arm is positioned in a mechanical arm positioner. Supplementary local anesthesia with epinephrine is injected in the area of the posterior portal for analgesia and hemostasis. The arthroscope is introduced through the posterior portal 2 cm below and 1 cm medial to the posterolateral corner of the acromion. The anterior working portal less than 1 cm lateral to the coracoid is localized using a spinal needle and established with a 5.5-mm cannula. Diagnostic arthroscopy then confirms the presence of a Bankart lesion. It is important to rule out associated pathology, including rotator cuff tears, anterosuperior labral tears, and posterior labral tears. Before committing to an arthroscopic repair, it is vital to assess for the presence or absence of bone loss. The bare area of the glenoid has been advocated as a useful landmark to assess for bone loss ( ) and while this technique may be utilized, it should be incoprorated in the overall assessment along with the preoperative imaging and Examination Under Anesthesia (EUA) because there have been recent publications demonstrating variability of this arthroscopic assessment tool ( ). If there is an off-track or engaging lesion with less than a 20% anterior glenoid bone defect, an arthroscopic stabilization is performed in conjunction with remplissage ( ); however, our threshold for moving to an open stabilization in this setting is low. Depending on the patient, sport, and position as well as the desired postoperative activity level, bipolar lesions with “subcritical” amounts of bone loss have recently been shown to have inferior outcomes after isolated Bankart repair. noted that in patients engaged in high-level overhead activity, glenoid bone loss greater than 13.5% led to inferior Western Ontario Shoulder Instability (WOSI) scores despite the lack of frank recurrent instability. Furthermore, in the setting of bipolar lesions, reported that glenoid bone loss of as little as 2 to 4 mm (8%–15%) in the setting of a moderate-sized Hill-Sachs lesion could compromise isolated Bankart repair. To that end, we carefully consider level and type and engage in shared decision making with patients when devising a surgical plan, which has resulted in a lower threshold for advanced arthroscopic and open techniques in the setting of bipolar lesions in high-demand patients. If there is an off-track or engaging lesion with a greater than 20% anterior glenoid defect, an open Latarjet reconstruction is routinely performed.

After confirming that there is no significant bone loss involving the glenoid or an engaging Hill-Sachs lesion to prevent successful arthroscopic repair, a second anteroinferior portal is placed under direct vision using a spinal needle. The angle of approach of this portal is critical: Before inserting an 8.25-mm clear, twist-in cannula, we use the spinal needle to be certain that the direction afforded by the portal is both inferior enough to reach the 5:30 position on the glenoid with an anchor and lateral enough to place anchors on the face of the glenoid rather than on the glenoid neck. In addition, ensuring an appropriate distance between the two anterior portal skin incisions (2–3 cm) facilitates instrument and suture management. At this time, we have found that a “bump” in the axilla in conjunction with the arm holding device can provide improved access and visualization of the inferior portion of the glenohumeral joint ( Fig. 1C.1 ).

Fig. 1C.1

Axillary bump. An axillary “bump” is used in conjunction with a pneumatic arm-holding device in order to provide hands-free glenohumeral distraction and subsequent improved access and visualization of the inferior portion of the joint.

From Moran CJ, Fabricant PD, Kang R, Cordasco FA. Arthroscopic double-row anterior stabilization and Bankart repair for the “high-risk” athlete. Arthrosc Tech. 2014;3(1):65-71.

After this second anterior portal is established, the labrum and capsule are mobilized from the glenoid using a combination of sharp elevator, dull elevator, and rasp. In an acute dislocation, the tissue is easily reducible, but in a chronic dislocator, the capsulolabral complex can be scarred down medially and inferiorly on the glenoid neck and must be mobilized for an anatomic repair. A 70-degree lens is helpful to visualize the anteroinferior and posteroinferior capsulolabral structures. The 70-degree lens is used for viewing via the posterior portal while elevating the labrum from the anterosuperior portal, especially when the labrum has healed in a medialized position. In this setting, we have found a sharp elevator ( Fig. 1C.2 ) to be useful in elevating scarred tissue as a single flap. The arthroscope may be placed in the anterosuperior portal for additional visualization.

Fig. 1C.2

Single-row repair using a knotless mattress suture configuration. This patient was a 24-year-old NFL football player who sustained an acute traumatic anterior glenohumeral instability event. Evaluation with a 30-degree arthroscope from the posterior portal revealed an anterior capsulolabral avulsion ( A ). An arthroscopic blade ( B ) is used to sharply elevate and mobilize the retracted capsulolabral tissue off the anterior glenoid followed by preparation of the glenoid neck with a rasp ( C ) and shaver ( D ) to stimulate bony bleeding. Nonabsorbable mattress sutures are placed sequentially, starting at the anteroinferior glenoid ( E ) and progressing superiorly ( F and G ) while viewing from the posterior portal with a 70-degree arthroscope. Knotless anchors are sequentially deployed ( H ) in a superior progression with attached mattress suture construct. Completed mattress repair with three knotless suture anchors ( I ).

Proper mobilization of the tissue is critical for arthroscopic stabilization; the glenoid labrum and capsule complex should be elevated to the 6 o’clock position to ensure restoration of the tension of these structures and closure of the axillary pouch. A probe is used to assess the mobility of tissue before placing any anchors. Another indicator that the capsule and labrum have been mobilized sufficiently is visualization of the subscapularis muscle under the mobilized capsule and labrum. After satisfactory mobilization of the labral and capsular tissue, the labrum should float up to its near anatomic position. Next, a motorized shaver is used to gently decorticate the glenoid neck to prepare a bleeding bed to aid in healing. This maneuver should be performed without shaver suction to prevent damaging the soft tissues.

It is important to intraoperatively differentiate anterior glenohumeral instability that results in capsulolabral tear, capsuloligamentous deformation, or a combination of both phenomena. During a dislocation event, the capsuloligamentous structures (e.g., IGHL) may undergo plastic deformation before frank failure at the capsulolabral junction, glenoid–labrum interface, or glenoid. In the setting of a small Bankart lesion and a greater degree of capsular laxity, greater plication may be performed without formal takedown of the capsular attachment by creating pleats in the capsule itself ( ). The pleats are performed first by gently abrading the capsular tissue to encourage vascularity and promote healing followed by grasping capsular tissue with a curved suture passer in two locations and either securing it to a suture anchor to complete the plication or more commonly incorporating the pleat in a knotless mattress construct secured to a knotless anchor. Increasing the space between the capsular bites increases the degree of plication and may be easily controlled by the surgeon. When performed in conjunction with Bankart repair, both sites of pathology may be addressed in the setting of a combined lesion. This may also be performed when laxity persists after labral repair ( ).

It is quite common for the capsulolabral damage to extend into the posteroinferior zone ( ). When this is noted, we believe it is imperative to include this area in the repair and reconstruction by switching portals and viewing via the anterosuperior portal and reattaching the posteroinferior capsulolabral tissue. This in effect restores the posterior attachment of the “hammock.” When we perform this associated repair, the anteroinferior sutures are not secured to the glenoid until the posteroinferior sutures have been placed.

Multiple suture configurations and techniques have been described for the treatment of glenohumeral instability, including single row, double row, traditional (knotted), and knotless. In general, we prefer knotless techniques to minimize irritating suture abrading the joint surfaces and synovium. Whereas single-row repairs are excellent at providing powerful capsular shift, double-row techniques restore a broad footprint anatomy. Both are described next.

Single-Row Repair Technique

We use a knotless anchor mattress technique ( Fig. 1C.2 ). The 70-degree lens while viewing posteriorly is particularly helpful at this stage. Capsular plication to address capsular stretch should be performed (depending on the athlete being treated) with the arm in 30 degrees of abduction and in 30 to 40 degrees of external rotation to prevent overtightening the anterior structures. The amount of capsule plicated can be managed by varying the amount of tissue included in the shift much like a “pleat” can be used with garments ( Fig. 1C.3 ). Suture-shuttling devices in the anteroinferior portal are used to pass stiff suture material through the capsule and into the defect between the labrum and the glenoid. The first suture is placed inferiorly. The limb is grasped and brought out through the anterosuperior portal. Without removing the shuttling device from the cannula, it is advanced through the capsule more superiorly and again advanced into the defect between the labrum and the glenoid. Here the suture is retrieved using a suture retriever and brought out through the anterosuperior portal. Before securing the suture to a knotless anchor, the next stitch is passed. Depending on the location of the capsulolabral injury, the repair can move from the most inferior 6 o’clock position as far superiorly as necessary. Care is taken to ensure that the suture anchors are placed 2 mm onto the glenoid face to prevent a medialized repair. Anchors placed more medial than this risk restoration of the glenoid labrum in a nonanatomic medialized position that will not restore the normal bumper effect of the anterior labrum. Conversely, if placed too far on the face of the glenoid, significant articular cartilage injury can occur.

Fig. 1C.3

Plication without formal takedown of capsular attachment. A suture passing device is used to secure capsular tissue (A) followed by a second pass (B) to perform a plication. The amount of capsular tightening can be managed by varying the amount of tissue included in the plication. The suture is subsequently tightened (C) and then secured with an arthroscopic knot.

Double-Row Repair Technique ( )

This technique is indicated for “high-risk” athletes (younger than 25 years of age, male, contact or forced overhead sports, higher levels of competition, more than two or three dislocations) and acute bony Bankart lesions that are amenable to arthroscopic repair ( ). We use a percutaneous long needle while staying lateral to the conjoined tendon. A guidewire and cannulated drill are introduced. The correct location for medial row anchors 10 to 15 mm medial to articular surface of glenoid is identified. Each of two to four anchors (depending on the size of the lesion) are placed without removing the drill guide, progressing from inferior to superior. The anchor suture limbs are passed from the medial row using outside-in or inside-out technique. Sutures are prepared on knotless anchors for lateral-row fixation at the articular margin. After an arthroscopic Bankart repair is completed, anterior stability is reassessed. The subacromial space is inspected to evaluate the bursal surface of the rotator cuff and coracoacromial arch.

  • Video 1C.1

    Technique for an arthroscopic double-row anterior stabilization. (From Moran CJ, Fabricant PD, Kang R, Cordasco FA. Arthroscopic double-row anterior stabilization and Bankart repair for the “high-risk” athlete. Arthrosc Tech . 2014;3(1):65-71.)


Improved instrumentation and understanding of the pathoanatomy and biomechanics of the shoulder have allowed surgeons to use advanced arthroscopic shoulder stabilization techniques in a larger fraction of patients with glenohumeral instability. Although patient position in either the beach-chair (modified Fowler) or lateral decubitus position may be used to successfully treat glenohumeral instability arthroscopically, the beach-chair position is preferred when the surgeon suspects pathology which may require conversion to an open approach (e.g., significant glenoid bone loss, poor capsulolabral tissue, HAGL lesion), which may be easily performed without patient repositioning.

Open Stabilization: Bankart and Capsular Shift and Humeral Avulsion of the Glenohumeral Ligament Repair

Troy A. Roberson, MD
Jared C. Bentley, MD
Richard J. Hawkins, MD


The open approach is considered the gold standard in addressing anterior instability, but there has been a trend toward more arthroscopic treatment of this pathology. There remains some concern that this trend may be due to a lack of familiarity with the open approach rather than a true advantage of arthroscopic stabilization. Despite similar overall results between the two approaches, some scenarios, including contact athletes, bone loss, and revision stabilization, may be particular indications for an open approach. Decision making for an open approach takes into account key history, examination, and imaging components because they help identify appropriately indicated patients. Examination under anesthesia is a key component of surgical planning because the degree of translation compared to the size of Bankart lesion is used to determine the degree of capsular plication required. In an isolated Bankart repair, the subscapularis and capsule are incised together, but in the case of a concomitant capsular shift, the capsule is elevated off the subscapularis. In a typical Bankart repair, three suture anchors are placed at the anterior glenoid rim beginning inferiorly and extending along the anteroinferior margin. The amount of capsular tissue incorporated into the repair depends on the amount of instability present on examination, the quality of the capsular tissue, and the size of the labral tear. In more extreme cases, a concomitant anterior capsular shift may be performed with a concern of restricted motion as a potential trade-off for increased stability in these difficult situations. In these cases, a cruciate capsulotomy is performed with advancement of the inferior leaflet and obliteration of the redundant inferior capsular space. Although arthroscopic stabilization of anterior shoulder instability is appropriate in many scenarios, open stabilization should not only remain a vital tool in shoulder surgeons’ armamentarium but in some cases may in fact be the preferred approach.

Keywords: anterior instability; Bankart; capsular shift.


The open approach has long been considered the gold standard in addressing anterior shoulder instability ( ).

  • Advances in arthroscopic technique over the past 2 decades have led to an increasing number of stabilization procedures being performed arthroscopically.

  • Multiple authors ( ) have noted a dramatic increase in arthroscopic stabilizations, with Zhang et al most recently reporting an increase in arthroscopic repair from 71% in 2004 to 89% in 2009.

  • In the United Kingdom, reported an increase in the preferred stabilization approach from 16% arthroscopic in 2002 to 71% in 2009. This is particularly interesting given that the comfort level with arthroscopic stabilization versus open procedures may be fueling this trend.

  • Considering much of these data were collected nearly a decade ago, the use of arthroscopic stabilization procedures may be considerably higher, especially among more recently trained surgeons who have had less exposure to open stabilization techniques.

Therefore it is reasonable to evaluate whether this increase is evidence based or simply an increased comfort level with the arthroscopic approach.

The controversy of open versus arthroscopic Bankart repair as evidenced by five previous meta-analyses ( ).

  • All found at least equivalent postoperative stability between arthroscopic and open repair.

  • Four ( ) of five found evidence of lower recurrent instability with open repair.

  • Three ( ) reported improved overall functional outcomes with open repair.

  • One ( ) found worse range of motion after open procedures.

There have been several retrospective comparative studies ( ) between open and arthroscopic treatment of anterior instability:

  • All found generally comparable recurrence rates and outcomes between the two.

  • The exceptions are one study ( ), which found a higher recurrence rate in arthroscopic repair but better range of motion versus open repair, and another reporting better postoperative functional outcomes after arthroscopic repair ( ).

Two nonrandomized prospective studies ( ) also report equivalent recurrence rates and outcomes.

  • do state that the open repair group was chosen based on the degree of capsular laxity with an additional capsular shift performed, indicating a distinction not only of the pathology treated but the difference in the nuances of the procedure.

Finally, four level I randomized controlled trials exist comparing these techniques.

  • Two ( ) report slightly better postoperative external rotation with arthroscopic treatment.

  • Two ( ) report significant decreases in recurrence rates with open versus arthroscopic repair.

  • The most recent study by reported a 23% recurrence rate in arthroscopic repair versus 11% in open repair with significant risk factors for recurrence being male gender under 25 years of age and significant preoperative Hill-Sachs lesion.

The totality of the presented literature suggests that open treatment of anterior shoulder instability may provide improved stability over arthroscopic treatment with the possible tradeoff of a modest decrease in range of motion in some cases.

There are some criticisms of open stabilization with presumed advantages of an arthroscopic approach.

  • Ease of operation: described operative time as an advantage of arthroscopic repair with open stabilization taking on average approximately 2.5 times longer. However, as we will show in the described technique, the open stabilization is not an onerous operation, and we believe this difference in operative time is more reflective of the previously mentioned lack of comfort with the open approach.

  • Concern of subscapularis failure: This may be exaggerated because subscapularis failure has been scarcely reported ( ).

    • Although found worse outcomes in patients with clinical signs of subscapularis insufficiency, only one of these patients had documentation by magnetic resonance imaging (MRI).

    • reported MRI evidence of atrophy of the upper subscapularis after an inverted L-shaped tenotomy, while the lower subscapularis remained intact with some evidence of compensatory hypertrophy, but patient outcomes were not reported based on the presence of radiographic subscapularis abnormality.

    • Despite these findings, two studies have shown no difference in internal rotation strength after arthroscopic versus open stabilization ( ).

    • Although it may be true that subscapularis dysfunction is a concern to some with the open approach, the available literature is inconclusive, and we have not found this to be a major concern in our practice.

  • Another historical negative of open stabilization is the loss of some degree of external rotation.

    • More contemporary approaches value this concern and allow management of the capsule in such a way to minimize the loss of motion without compromising stability.

  • An open approach allows uncomplicated treatment of the full pathology encountered including humeral avulsion of the glenohumeral ligament (HAGL) lesions, which are an uncommon but real cause of recurrent instability ( ).

With these concerns mitigated and results that meet or exceed those of arthroscopic repair, we suggest that the trend toward arthroscopic stabilization is based more on comfort level than other factors, and open stabilization should not only remain a vital tool in the armamentarium of shoulder surgeons but in some cases may be the preferred technique. In the remainder of the chapter, we will discuss when this procedure may be particularly indicated or contraindicated, the key history and physical examination findings to promote success, and our preferred surgical technique.

Indications and Contraindications

Although successful surgery may be achieved in many instances with either an arthroscopic or open approach at discretion of the surgeon, certain situations may be indicated more strongly for open stabilization. Contact athletes provide one of the most challenging patient populations in shoulder instability treatment. Results of arthroscopic stabilization have generally been discouraging with failure rates ranging from 14% to 60% ( ). found a 60% recurrence rate after arthroscopic stabilization versus 0% with open stabilization, and reported twice the risk of recurrence (25% vs 12.5%) with arthroscopic stabilization in this population. In contrast, reported only a 3% recurrence rate among American football players with open stabilization. When taking age into consideration, open stabilization may be even more attractive. Only 57% of American Shoulder and Elbow Surgeon (ASES) member respondents chose arthroscopic stabilization as their preferred approach to the young contact athlete ( ). In this historically troublesome population, recently reported no recurrences and good to excellent outcome scores in all patients at greater than 2-year follow-up with open stabilization.

Overlapping with the discussion of contact athletes, highlighted the importance of bone loss in recurrent instability, reporting a 89% failure rate in contact athletes with significant bone loss about the glenoid or humeral head and 67% recurrence in the overall group in the presence of bone loss with arthroscopic stabilization. With this concern over bone loss, the rate of bone block procedures has increased over the past 2 decades ( ). and some have advocated for its place as a primary procedure in anterior instability ( ). However, multiple authors have reported complication rates exceeding 20%, leaving understandable concern whether this should be considered a first-line treatment ( ). An option in between arthroscopic stabilization and bone block procedures may be open stabilization, but few studies have examined the results of open stabilization in the face of bone loss. In their classic study, found a slight decrease in recurrence with glenoid bone loss (3.5% vs 2%) and a slight increase in failure with moderate to severe Hill-Sachs lesions (3.5% vs 5%). Whereas reported an overall 12% recurrence with varying degrees of glenoid rim defects with a technique that included capsular shift, while found an increase in recurrence from 3% to 6% in the presence of a large Hill-Sachs lesion with open stabilization. More recently, reported similar results with low recurrence rates in the face of glenoid or humeral head bone loss with open stabilization. These results suggest that the concerns over bone loss in arthroscopic stabilization may not be applicable to open stabilization, and we suggest open stabilization may be indicated and more advantageous in many instances when others propose bone block procedures.

Another instance in which open stabilization may be useful is in the situation of revision operations. Although arthroscopic procedures may be considered for revision surgeries, open stabilization has shown good results with recurrent instability reported from 0% to 22% ( ). Most recently, reported no recurrences with excellent patient reported outcomes and return to sport at a mean of over 10 years from revision open stabilization. In addition to those above, a primary failure of the subscapularis or concomitant HAGL lesion may be easily addressed with an open approach.

There are few contraindications to open stabilization with voluntary instability or psychological disease among them ( ). Large defects of the humeral head or glenoid may require bony augmentation, but as we have presented, most cases including bone loss may be successfully treated with a standard open stabilization. Finally, overhead throwing athletes are often preferentially treated with arthroscopic stabilization due to gravity of any issue with subscapularis dysfunction or loss of external rotation.


A thorough history and physical examination is the cornerstone of diagnosing and treating patients with shoulder instability. Among the key questions are age, handedness, initial or multiple events, number of previous events, amount of trauma required for instability, voluntary dislocation, position of the arm, whether Emergency Department (ED) reduction was required, whether it was painful, if there was dislocation during sleep, and what sport or recreational activities are desired. Of these, one of the most critical is age because it has been defined as one of the most important risk factors for higher recurrence with nonoperative management ( ). On the other end of the spectrum, patients older than 40 years of age have lower recurrence rates but should be closely scrutinized for a concomitant rotator cuff tear. Patients should be questioned to elicit signs of generalized ligamentous laxity and thus the possibility of multidirectional instability as opposed to isolated anterior instability. A first time dislocator or one who has had multiple dislocations requiring ED reduction is quite different from a patient who describes self-reduced events, but it is possible for a multidirectional lax individual to experience an isolated anterior instability event. The presence and distribution of pain should also be considered because neck pain or radiating arm pain may indicate cervical spine involvement or brachial plexus pathology. Pain in the subdeltoid region associated with worsening at night causes concern for rotator cuff involvement. A patient’s desired activity level is one of the final important considerations, with contact athletes and overhead athletes getting special consideration. In the tertiary care setting, many patients present with a prior surgical history, which should be probed in detail with confirmation of previous medical records.

Physical Examination

Although a full examination of the shoulder should be performed to assess contour, range of motion, strength, and for the presence of neurovascular abnormalities, we provide a focused discussion of specific tests to aid in treatment decisions. The sulcus sign is assessed in the upright position with an inferiorly directed force on the arm with grade 1 being less than a 1-cm gap between the acromion and humeral head, grade 2 between 1 and 2 cm, and grade 3 more than 2 cm. It is critical to ask the patient whether this translation reproduces the symptoms or is simply a physiologic laxity. In the supine position, the anterior apprehension test, or fulcrum test, is performed in 90 degrees of abduction with gentle external rotation of the arm ( Fig. 1D.1 ). As the arm is rotated, a positive test result is elicited with apprehension or feeling of a near instability event by the patient or examiner. A relocation test may be performed at this point with posteriorly directed force on the humeral head with a positive response being relief of the apprehension ( Fig. 1D.2 ). Some may associate a relocation test as relief of pain with the same maneuver; however, this response is consistent with a diagnosis of internal impingement, not anterior instability. Although other tests may be performed for concomitant pathology, the final test specifically for instability is the load and shift ( Fig. 1D.3 ). With an axial load, the humeral head is translated anterior with grade 1 being translation to the edge of the glenoid, grade 2 requiring translation over the glenoid rim but spontaneous reduction, and grade 3 involving translation beyond the glenoid rim that requires manual reduction ( Fig. 1D.4 ). It is critical to assess whether these tests result in a reproduction of symptoms, and they combine to give a picture of the degree of instability and its correlation to symptoms.

Fig. 1D.1

Performance of the anterior apprehension test. An examination result is positive when the patient expresses apprehension or the feeling of the shoulder slipping out of the socket when in this abducted externally rotated position.

Reprinted from Romeo A, Provencher M, eds. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2016.

Fig. 1D.2

This clinical photograph shows the performance of the anterior relocation test. The addition of a posteriorly directed force to the anterior proximal arm results in relief of patient apprehension in the abducted, externally rotated position.

Reprinted from Romeo A, Provencher M, eds. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2016.

Fig. 1D.3

A and B, Clinical photographs showing the performance of the anterior load-shift examination. The examiner loads the humerus longitudinally into the glenoid socket and attempts to translate the humeral head over the glenoid rim in the anterior direction. This test also may be performed in the posterior and inferior directions.

Reprinted from Romeo A, Provencher M, eds. Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2016.

Fig. 1D.4

Load-and-shift test. The adequacy of the glenoid concavity in a given direction can be assessed by compressing the humeral head into the glenoid concavity and noting the amount of displacing force necessary to translate the head. A shows a normal concavity, and B shows a diminished concavity in the direction of texting.

From Matsen FA, III, et al: Shoulder Surgery: Principles and Procedures . Philadelphia: WB Saunders; 2004:113.


Plain radiographs remain the first-line imaging with standard anteroposterior (AP), scapular Y, and axillary views most commonly used. The scapular Y may be used to assess the position of the humeral head, but the axillary view is required to confirm position and reduction, especially in acute cases. The West Point view may be useful in evaluating the glenoid rim, and the Stryker notch view is helpful in assessing the extent of Hill-Sachs lesions ( ). In the case of a clear diagnosis of anterior instability without red flags for significant bone loss (i.e., plain radiograph findings, apprehension or instability at lower levels of abduction, low-energy dislocations), these plain radiographs may be sufficient. However, in many cases, advanced imaging such as computed tomography (CT) or MRI is advantageous in surgical planning. MRI has progressed to the point of excellent accuracy in confirming the diagnosis of injury to the capsulolabral complex ( ). Although CT has generally been the modality of choice in assessing bone loss and three-dimensional CT at this point is considered the most reliable ( ), more recently, MRI has been shown to be an acceptable tool ( ) for this purpose and may obviate the need for multiple three-dimensional scans.

Technique ( )

Every patient undergoing an instability procedure is routinely examined under anesthesia. The degree and direction of humeral translation as well as range of motion are objectively quantified and compared with the contralateral side. This step is vital to assisting in intraoperative decision making because the amount of capsular plication must take into consideration the degree of translation and size of the Bankart lesion (i.e., large translation on examination with small lesion present intraoperatively calls for a more substantial plication or shift). After the examination, the patient is placed in the beach-chair position with the head of the operating table raised 30 degrees. A bump is placed to stabilize the scapula, and the shoulder is prepped and draped with the arm free.

  • Video 1D.1

    Surgical procedure.

A vertical anterior axillary incision is used and can be identified preemptively by digitally pulling the skin vertically at the coracoid to identify an axillary crease. The incision is carried from directly inferior to the coracoid to the axillary margin. The incision can also be shifted inferiorly into the axilla in female patients for improved cosmesis but does require more superior retraction. Deeper dissection is carried through deltopectoral interval after identification of the cephalic vein, which is most often taken laterally with the deltoid. A Goelet or Richardson retractor is used to retract the deltoid laterally. The clavipectoral fascia is incised just lateral to the “red stripe” or muscular portion of the coracobrachialis and excised to permit visualization. In more muscular patients, the falciform fascial expansion of the pectoralis major tendon can be incised 1 cm inferiorly to assist in exposure. A Hohmann retractor is placed under coracoacromial ligament. Subacromial and subdeltoid tissue adhesions are freed using a Cobb elevator to facilitate placement of a Brown deltoid or self-retaining retractor. The short flexors and conjoined tendon are then freed digitally and then bluntly retracted, medially exposing the underlying subscapularis while being cognizant to limit prolonged forceful retraction distally because of the musculocutaneous nerve. The Coracoacromial (CA) ligament can be partially incised to assist in mobilizing the short flexors, further aiding visualization. The axillary and often the musculocutaneous nerves can be palpated on the undersurface of the conjoined tendon. The arm is externally rotated, fully exposing the subscapularis tendon. The biceps tendon and rotator interval are visually and palpably identified and serve as the barometer for preparation for the subscapularis tenotomy.

When an isolated Bankart repair is being performed, the subscapularis and underlying capsule are incised and elevated in continuity as one capsulotendinous unit ( Fig. 1D.5 ). The tenotomy is performed beginning at the rotator interval superiorly and carried distally approximately 1.5 cm medial to the bicep tendon with the arm abducted and in external rotation to distance the tenotomy from the axillary nerve. It is important to leave a good cuff of tissue laterally on the greater tuberosity for later subscapularis repair. The lower third of the subscapularis is incised slightly more medially as the insertion becomes more muscular. Discrete branches of the circumflex vessels are coagulated with electrocautery or ligated with silk suture. Two heavy braided stay sutures are placed through the tendon and used for retraction as well as later repair ( Fig. 1D.5 ). Metzenbaum scissors are used to incise the interval from lateral to medial. The humerus is translated laterally, and a Carter-Rowe or Fukuda humeral head retractor is then placed, exposing the glenoid and anterior capsulolabral structures. The blunt right-angled retractor previously used for retracting the conjoined tendon medially is replaced intra-articularly, reflecting the capsular-tendinous flap medially. The Bankart lesion is then clearly visualized and mobilized with a series of elevators exposing the underlying anterior glenoid bone. The mobilization of the lesion is complete when the undersurface of the subscapularis is visible within the lesion medially. In larger lesions, a cleated anterior glenoid retractor can be placed inside the labral lesion to facilitate exposure of the bone bed. The bony anterior glenoid is then prepared biologically using manually powered rasps or curettage. In the typical Bankart lesion, three resorbable 3.0-mm single-loaded anchors with nonabsorbable stitch (SutureTak; Arthrex, Naples, FL) are placed beginning as low as the 5:30 or 6 o’clock position inferiorly. These anchors are placed directly onto the anterior margin glenoid face with attention paid to avoid medialization of the repair. Following anchor placement, the anterior glenoid retractor is removed. A tapered free needle or arthroscopic suture shuttle (Suture Lasso; Arthrex) is used to pass suture through the capsulolabral tissue. The amount of capsular tissue incorporated into the repair depends on the amount of instability present on examination, the quality of the capsular tissue, and the size of the labral tear. In general, a 5- to 7-mm capsular plication is performed; however, too-aggressive plication will place undue tension on the subscapularis repair later in the operation ( Fig. 1D.6 ). Simple stitches are then tied, although one could consider mattress configuration by passing both limbs of the anchor through the capsule, thereby placing the knots extraarticular. This completes the Bankart repair. The subscapularis is then repaired using #5 braided nonabsorbable stitch in Mason Allen fashion from proximal to distal, typically using the two sutures that previously passed initially and o additional sutures. The degree of interval closure is titrated to the degree of stability gained and quality of tissue at the repair and is most often executed with two sutures incorporating 2 to 3 mm of interval and subscapularis tissue passed in figure-of-eight fashion. A standard layered interval and skin closure follows. Drains are typically not used.

Fig. 1D.5

Cadaveric specimen, right shoulder. When an open Bankart repair is conducted without concomitant shift, the subscapularis and capsule are elevated as one and tagged with heavy nonabsorbable suture for retraction and later repair.

Fig. 1D.6

Cadaveric specimen, left shoulder. Inferior and superior leaflets of the laterally based T-capsulotomy are tagged in preparation for advancement of the inferior flap, obliterating the anterior inferior capsular redundancy.

In more complex anterior instability cases, including those involving glenoid bone loss of more than 10% to 15%, poor capsulolabral tissue, history of numerous dislocations, or patients with multidirectional laxity, an anterior capsular shift may be performed with Bankart repair. This repair represents two procedures with some concern for “overtightening” and increased potential for restricted motion. Performing a concomitant capsular shift requires several deviations from the described technique. After isolation of the subscapularis, but prior to tenotomy, the inferior lateral margin of the subscapularis is incised transversely, just superior to transverse humeral vessels. Metzenbaum scissors are then placed into the incision, under the subscapularis, and advanced superiorly into the interval bluntly separating capsule from subscapularis tendon. A tenotomy is then performed directly on top of the scissors allowing a sufficient lateral cuff of tissue for later tendinous repair. Nonabsorbable heavy braided retention stitches are used for retraction, and the subscapularis is carefully, bluntly dissected inferolaterally to superomedially with scissors.

With the subscapularis reflected and capsule exposed, a laterally based cruciate or “T” capsulotomy is performed. Laterally, the capsular incision is carried vertically to the rotator interval. The inferior leaflet is dissected off the humeral neck inferiorly to permit appropriate obliteration of the anterior-inferior pouch when advanced. The transverse limb of the cruciate capsular incision is made at the midglenoid level and extended medially to the labral margin. Superior and inferior leaflets are tagged laterally and retracted, and the Bankart repair is conducted as described earlier, with the “north–south” capsular shift performed after completion (see Fig. 1D.6 ). Adequate capsular release and mobilization of the leaflets can be appreciated palpably by placing a digit into the inferior pouch, which should collapse when the leaflet is advanced, or visibly as the exposed humeral articular surface should be nearly covered with simulated shift ( Fig. 1D.7 ). The rotator interval is then closed laterally to form a firm foundation to base or “hang” the inferior shift. The inferior capsular leaflet is then mobilized superiorly. In preparation for capsular repair, the shoulder is placed in 30 degrees of abduction and external rotation. The first stitch is a corner stitch that will “set” the degree of shift and is followed by three additional mattress sutures inferiorly for additional security. The superior reconstruction is conducted by taking the superior leaflet inferiorly, over the top of the inferior flap, and securing it laterally with two sutures. Careful attention is paid not to suture the superior leaflet to its inferior counterpart because this can cause increased tension at the inferior capsular repair. The subscapularis is then repaired as described earlier.

Fig. 1D.7

Cadaveric specimen, right shoulder. A capsulorrhaphy is incorporated into the labral repair and must be tailored to the patient.

Although less commonly encountered, HAGL lesions can be effectively managed using the traditional open techniques. An open, trans-subscapularis technique is used in cases of anterior lesions. The standard deltopectoral approach to expose the subscapularis is conducted as described earlier; however, the axillary incision can be made smaller using only the inferior 5 cm. Through this incision, the subcutaneous tissues are elevated from underlying fascia to increase mobility of the subcutaneous “window” as the interval outlined by the bicep tendon sheath laterally, subscapularis tendon medially, and pectoralis major tendon inferior is identified. Blunt dissection scissors are then used to establish a tissue plane between the capsule and subscapularis. This is done by penetrating the subscapularis in line with its fibers just superior and medial to the circumflex vessels, advancing the scissors superiorly approximately 2 cm, and then opening the blades creating a clear soft tissue plane. An L-shaped subscapularis tenotomy is performed, making a vertically subscapularis tendon incision 1.5 cm medial to the bicep tendon, and is directed from the midtendon inferiorly to just superior to the circumflex vessels. The incision is then taken from the inferior margin laterally to complete the “L.” A stitch is placed in the subscapularis, allowing retraction and facilitating further separation from the underlying capsule. The humerus is then externally rotated and abducted to identify the avulsed capsule. The axillary nerve is identified medially by palpation and protected. After the lesion has been fully exposed, a direct repair is done using 3-mm anchors double loaded with braided nonabsorbable suture. Arthroscopic suture-passing instruments can be very helpful when working in the confined space of the axillary pouch. After repair, a standard tenotomy repair and closure is performed.


In general, patients undergoing open stabilization for anterior/inferior instability are placed in a sling secured anteriorly across the abdomen positioning the shoulder in neutral to slight internal rotation. Passive range of motion is initiated immediately, restricting supine external rotation to neutral for the first 2 weeks and then increasing to 30 degrees from postoperative weeks 2 to 4. After week 4, the sling is weaned, active range of motion is initiated, and external rotation is gradually increased to 45 degrees by week 8. Resistance training is initiated 6 weeks postoperatively, including internal and external rotation strengthening with bands and manual resistance scapular stabilization drills. Weeks 8 to 12 focus on regaining terminal external rotation and transitioning to restricted weight training. General restrictions focus on avoidance of anterior capsular stress and include keeping the hands within eyesight; keeping the elbows bent; and minimizing overhead activities such as military press, pull-downs, and so on. Conventional weight training commences at week 16. Reintroduction to sport generally begins at 4 months in both contact and overhead athletes with full return by 6 months.


Recurrent instability is of most concern, is well documented, and the most commonly cited complication after shoulder stabilization procedures. Recurrence after open repair is 5% or less in the vast majority of studies. Several series have shown similarly low rates of recurrence in collision athletes undergoing open procedures. Additionally, a meta-analysis including 18 studies reported a higher risk of recurrent instability (18% vs 8%) and reoperation Relative Risk ((RR), 2.32) in patients undergoing arthroscopic stabilization procedures compared with open ( ).

Subscapularis failure and secondary dysfunction are complications inherent to open stabilization procedures and a clear distinction from arthroscopic repairs. Rates of reoperation for subscapularis rupture after tenotomy for open stabilization have been reported as high as 4.5% ( , ). However, numerous additional series do not report any revisions caused by postoperative subscapularis tendon disruption. Conversely, motion restriction and stiffness can occur. However, reported no patient lost more than 15 degrees in a series of 58 professional athletes. The risks of postoperative subscapularis failure or stiffness highlight the importance of tailoring the capsulorrhaphy to the patient’s clinical examination, meticulous surgical technique, and subscapularis repair as well as carefully monitored rehabilitation.

Damage to neurovascular structures and secondary dysfunction, although rare, have been reported. The axillary nerve must be identified and protected throughout the procedure. External rotation and abduction of the humerus during exposure and dissection increase the distance of the nerve from the surgical field. If self-retaining retractors are used on the conjoined tendon, they should be kept proximal avoiding excessive tension and prolonged placement to prevent undue traction on the musculocutaneous nerve. Wound complications are rare, and the vertical axillary incision, in our experience, has been cosmetically favorable.


Open Bankart repair and anterior stabilization procedures in the care of injured athletes’ shoulders remains a powerful instrument in shoulder surgeons’ toolbox. Their role is critically important in young collision athletes whose risk for recurrent instability and its implications are the highest. The literature supports open techniques for stabilization in contact athletes because the outcomes appear superior to arthroscopic techniques in this high-risk subset. Although complications inherent to open shoulder procedures exist, they can be mitigated with refined technique and appropriate rehabilitation. Despite arthroscopic advances, there lies a clear place for open capsulolabral reconstruction in the shoulder surgeons’ armamentarium for management of anterior instability.

Glenoid Bone Loss: Open Latarjet

Matthew T. Provencher, MD, CAPT, MC, USNR
Anthony Sanchez, BS
George Sanchez, BS
Tistia Gaston, PA-C


First described in 1954, the open Latarjet procedure is undertaken for treatment of recurrent anterior glenohumeral joint instability. The procedure features fixation of the coracoid process with its conjoined tendon to the glenoid rim along with management of the subscapularis and joint capsule. Regardless of many alterations in technique, this procedure remains the “gold standard,” particularly in cases of clinically relevant glenoid bone loss. Here we present an overview of our preferred technique as well as the indications, necessary workup, key technical pearls, and potential complications to take into account for the open Latarjet procedure. Last, we compare the open Latarjet procedure with the all-arthroscopic Latarjet procedure.

Keywords: anterior shoulder instability; bony augmentation; coracoid process; glenoid bone loss; Latarjet.


  • Glenoid bone loss is a commonly encountered contributing pathology to recurrent anterior glenohumeral instability.

  • The open Latarjet procedure is an established, validated technique that addresses this pathology via reconstruction of the bony defect and reinforcement of the surrounding associated supporting tissues, restoring stability.

  • Proper execution of this technique starts with patient selection informed by physical examination and imaging studies and depends on appropriate coracoid process harvest, transfer, and fixation. Procedure outcomes are ensured by postoperative rehabilitation focused on graft union optimization.

Patient Selection


The Latarjet procedure is indicated for patients with recurrent anterior instability, significant loss in the glenoid or humeral bone, and failed prior Bankart repairs. In particular, the Bankart failure rate has been shown to increase from 4% to 67% with significant bone loss, specifically defined as greater than 25% loss of the inferior glenoid diameter giving an “inverted pear” appearance, wherein the inferior glenoid has a smaller diameter than the superior glenoid ( ). The incidence rates of bone defects in cases of chronic anterior shoulder instability have been radiographically identified to include a bony lesion rate, humeral or glenoid, of 95% with a humeral impaction fracture rate of 73.1% ( ). Patients with an inverted-pear glenoid have been previously reported to have a 67% failure rate after Bankart repair versus a 4.9% recurrence rate after open modified Latarjet.

Additionally, it is important to recognize the limitations of the Latarjet procedure and thus when it is contraindicated. When more than one third of the articular surface of the anterior glenoid is fractured or depleted, the coracoid is not sufficient. In these cases and after a failed Latarjet procedure, reconstruction is performed with an iliac crest bone graft or distal tibial allograft ( ).

Physical Examination

As part of a standardized preoperative physical examination to rule out other pathologies, range of motion, rotator cuff, neurovascular status, and posterior instability are tested in addition to anterior stability. Using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) tool, the diagnostic accuracy of individual anterior instability physical examination tests was measured ( ). In descending order from high to low diagnostic accuracy the available tests for assessing anterior instability ranked as follows: relocation or Jobe relocation test ( ), Apprehension ( ), anterior release/surprise test ( ), anterior drawer test ( ), and anterior-superior instability with an anterior supraspinatus tear (SLAC lesion) tests (load and shift test, anterior-superior SLAP test, active compression or O’Brien test, Kibler test, and Whipple test [ ]).

Key Clues from History and Physical Examination That Suggest a Diagnosis of Glenoid Bone Loss

During assessment of a patient with recurrent anterior instability and possible glenoid bone loss, certain key clues will allude to a diagnosis of bone loss. These clues should be sought and considered for a definitive diagnosis. In regard to history, the patient should report a prolonged history of instability as well as a previous high-energy mechanism of injury. Most important, the patient should note most instability occurs in midrange of motion (20–60 degrees of abduction) because this highly suggests bone loss. During physical examination, shoulder apprehension should be very apparent in midranges of abduction (30–90 degrees) and less apparent in external rotation. Last, anterior translation of the humeral head over the glenoid rim should be reproducible ( ).

Preoperative Imaging and Arthroscopic Validation

The main method for detection of glenoid bone loss is the circle method on three dimensionally reconstructed computed tomography described by This technique consists of performing imaging with three-dimensional (3D) computed tomography (CT) ( ), which has been demonstrated to be the most reliable imaging modality, and subsequently using processing software to obtain the en face image of the glenoid fossa with the humeral head digitally subtracted in order to quantify bone loss. The area of the displayed osseous fragment or erosion traced by the particular software over the area of a best-fit circle based on the inferior two thirds glenoid contour from 3 to 9 o’clock is used to calculate total bone loss ( ). The glenoid bone loss quantification measure obtained by the circle method can then be reaffirmed arthroscopically. An arthroscope is used through the anterosuperior portal while a calibrated probe is inserted via the posterior portal to measure the distance from the anterior and posterior rims to the bare spot ( ). Despite questions as to bare spot landmark consistency, arthroscopic measurement is still well documented and is routine when planning an open Latarjet procedure ( ).

Preferred Technique for Measurement of Glenoid Bone Loss

For measurement of glenoid bone loss, the senior author (MTP) prefers the use of the ratio method, which uses the intersection between the longitudinal axis and the widest anteroposterior diameter of the glenoid in an en face view of the glenoid on CT scan ( ). With the bare area approximated, a best-fit circle is centered on the inferior two thirds of the glenoid. The distance from the bare spot to the edge of the defect is measured (d) as well as the radius of the best-fit circle (R). The ratio d/R is then inserted into the equation shown in Fig. 1E.1 for quantification of percent bone loss ( ).

Fig. 1E.1

Measurement of glenoid bone loss based on ratios. An en face view of the glenoid is acquired via computed tomography scan. At the intersection of the longitudinal axis and the widest anteroposterior diameter of the glenoid, the bare area is noted. A best-fit circle centered at the bare spot is then drawn about the inferior two thirds of the glenoid (dashed red line). The distance to the anterior edge (d) is measured and compared with the radius (R). The ratio of d/R is entered into the equation shown for quantification of percent bone loss.

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Preferred Surgical Technique

Positioning and Preparation

After interscalene block placement and general anesthesia induction, the patient is placed in a modified beach-chair position with the head of the bed elevated 40 degrees ( ). Two folded towels are placed under the medial border of the ipsilateral scapula to flatten and stabilize the scapula. The arm is draped free to allow for intraoperative abduction and external rotation. To prevent the arm from dangling and shifting when not being manipulated, a pneumatic limb positioner (Smith & Nephew, Andover, MA) or a padded Mayo stand is used.

Exposure and Approach

An arthroscopic confirmation of diagnosis and extent of osseous deficiency is performed, and any concurrent intraarticular pathology may be treated at this point. An oblique 5- to 7-cm incision is made from the tip of the coracoid process and extending inferiorly down the deltopectoral groove to the superior portion of the axillary fold ( ). While maintaining hemostasis, a standard deltopectoral approach is used. The cephalic vein and its consistent medial branches are protected and gently retracted laterally with the deltoid musculature ( ). A self-retaining Kolbel retractor is placed between the pectoralis major and deltoid to maintain exposure. If more exposure is desired, a Hohmann retractor over the top of the coracoid may be used while the arm is in abduction and external rotation.

Coracoid Graft Harvest

With proper coracoid exposure and the arm in external rotation and abduction, Mayo scissors are used to further expose the coracoid from its tip all the way to the insertion of the coracoclavicular ligaments at the coracoid base. The CA ligament is identified and sharply transected 1 cm laterally off its coracoid insertion ( Fig. 1E.2 ). It is important to harvest 1 cm of this ligament so it can later be incorporated into the capsular repair to yield the bumper effect. To improve exposure on the medial side of the coracoid, the arm is now placed in adduction and internal rotation. The pectoralis minor is released directly from the coracoid by taking down its insertion sharply with an elevator, with care to protect the neurovascular structures inferiorly with a blunt retractor. The release should not continue past the tip of the coracoid to avoid risk to the graft blood supply. Using a periosteal elevator, excess soft tissue is removed from the undersurface of the coracoid. Palpation to identify and protect the axillary and musculocutaneous nerves is necessary throughout the coracoid exposure.

Fig. 1E.2

The coracoacromial ligament is incised 1 cm from the coracoid insertion (arm in externally rotated position). Note the position of the Hohmann retractor over the top of the coracoid process.

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Using a 90-degree oscillating saw blade, a medial-to-lateral osteotomy of the coracoid is created at a line just anterior to the coracoclavicular ligament insertion at the coracoid base ( ). Ensure that the coracoid graft will be 22 to 25 mm in length from the tip using a ruler ( Fig. 1E.3 ). Make sure that the osteotomy is done perpendicular to the coracoid process to avoid accidentally extending it to the glenoid articular surface. An angled saw is used instead of a half-inch osteotome because the saw is less likely to cause iatrogenic glenoid fracture. Levering on the fragment with an osteotome can assist in completing the osteotomy but should be avoided if possible to avoid splitting the fragment. Chandler elevators are positioned inferior and medial to the coracoid to protect vital neurovascular structures. The blood supply to the graft enters the coracoid at the medial aspect of the insertion of the conjoined tendon; care is taken not to disturb it while performing the osteotomy. After the osteotomy, toothed grasping forceps are used to gently hold the graft at the level of the incision, and the coracohumeral ligament is released to liberate the coracoid.

Fig. 1E.3

A, The coracoid is first exposed to carry out a successful osteotomy. B, After release of the pectoralis minor and preparation of the inferior aspect of the coracoid with a periosteal elevator, the osteotomy is performed at the “knee” of the coracoid (i.e., distal to the coracoclavicular ligaments) (arm in adducted, internally rotated position).

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

The musculocutaneous nerve is then identified and released from the posterior fascia of the conjoined tendon, just until it dives into muscle approximately 4 to 7 cm from the coracoid tip. The coracoid is brought out of the incision more than 1 to 2 cm to avoid any tension on the musculocutaneous nerve. It is paramount to completely release all soft tissue adhesion on the posterior aspect of the conjoint tendon to allow for ease of coracoid transfer. Moreover, the musculocutaneous nerve is also visualized and gently released until all tension is freed all the way up to the musculocutaneous nerve insertion into the muscle. The arm is now returned to the neutral position.

Coracoid Graft Preparation

Using a scalpel, any remaining soft tissue is removed from the deep surfaces of the graft, while making sure not to harm the blood supply or CA ligament stump. A microsagittal saw is then used to decorticate the medial coracoid surface and expose a broad, flat cancellous surface to optimize graft union. If the osteotomy resulted in a spike of bone (which means optimal length harvest), it is removed while making sure not to compromise the length of the graft. An osteotome is then placed beneath the coracoid, and a 3.2-mm drill is used to place two bicortical drill holes along the central longitudinal axis of the graft about 1 cm apart ( Fig. 1E.4 ).

Fig. 1E.4

Preparation of the coracoid process. A gauze sponge is used to protect the skin.

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Glenoid Exposure and Preparation

The arm is externally rotated, putting the subscapularis on stretch. The subscapularis superior and inferior margins are identified, and the subscapularis split is performed with a #15 blade, 3 to 4 mm medial to the medial aspect of the biceps tendon. It is then divided medially up to the level of the coracoid but not farther because the subscapularis nerve supply enters the subscapularis about 1 to 1.5 cm medial to the coracoid. As first described by , our preference is to split the subscapularis at the junction of the superior two thirds and inferior one third rather than detaching distally at the insertion. Then, by pushing the scissors as far as the capsule and opening them perpendicular to the direction of the muscle fibers, the plane between the upper subscapularis and anterior capsule is developed and better visualized. Next, a single-prong self-retaining Gelpi retractor is used to open the subscapularis split ( Fig. 1E.5 ). The capsular cut is extended with a scalpel to the lesser tuberosity for exposure of the glenohumeral joint line and capsule. To preserve length, an L-shaped capsulotomy is performed—superior first at the superior glenoid and approximately 1 cm medial to the glenoid rim. A #2 high-strength suture is then used to tag the corner (superomedial) of the capsule for later repair. Electrocautery is next used to excise the anteroinferior labrum and periosteum off the region of the glenoid neck where the coracoid graft will sit. Last, a high-speed bur is used to lightly decorticate the anterior glenoid neck and create a flat, bleeding cancellous surface for graft placement. It is essential to form a bleeding bed of bone on the anterior glenoid and prepare the anterior aspect of the glenoid as perpendicular to the surface as possible.

Fig. 1E.5

A, Division of the subscapularis at the junction of the upper two thirds and inferior one third. Arrows depict the extesnion of the subscapularis split to the level of the coracoid. B, The subscapularis is split in line with its fibers; then the scissors are oriented vertically to expose the underlying capsule, and a sponge is placed superomedially. Arrows depict opening of scissors to develop plane between capsule and subscapularis.

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Coracoid Graft Fixation

Proper coracoid placement is arguably the most critical aspect of the Latarjet procedure. noted that excessive lateral coracoid placement yields an elevated postoperative rate of degenerative changes. Additionally, failure to correct the recurrent anterior instability will occur if the graft is overly medialized. The graft should function as an extension of the previously deficient, inherent articular arc of the glenoid. A low-profile Fukuda retractor is inserted inside the joint to retract the humeral head. Exposure can be improved superiorly by placing a 4-mm Steinmann pin into the superior scapular neck, medially by replacing the Hohmann from earlier with the link Hohmann retractor, and inferiorly by placing the same Hohmann between the capsule and the subscapularis, thereby exposing the 6 o’clock position on the glenoid.

The original traditional Latarjet procedure first described in 1954 called for fixation of the horizontal limb of the coracoid process with a screw flush to the anteroinferior margin of the glenoid, after having made a horizontal incision through the fibers of the subscapularis ( ). In this technique, the lateral surface of the coracoid becomes the face of the glenoid ( Fig. 1E.6 ). later proposed a modification to the procedure consisting of suturing the anterior joint capsule to the stump of the coracoacromial ligament.

Fig. 1E.6

Removal of the coracoid process for coracoid graft harvest and placement of the graft with its inferior surface directly apposing the glenoid rim so that the lateral surface of the coracoid becomes the face of the glenoid as done in the traditional Latarjet procedure.

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

de Beer et al first described the congruent-arc modification to the traditional Latarjet procedure. In this modification, the coracoid is rotated about its axis by 90 degrees to lay the inferior surface of the coracoid, which they found to be identical in radius of curvature to that of the native glenoid surface, parallel to the surface of the glenoid, thus eliminating the need to conform the graft with a bur ( Fig. 1E.7 ) ( ). Essentially, this rotated position makes it so that the inferior (deep) coracoid surface becomes contiguous with the glenoid surface ( ).

Fig. 1E.7

Removal of the coracoid process for coracoid graft harvest and placement of the graft with its lateral surface directly apposing the glenoid rim so that the inferior surface of the coracoid becomes the face of the glenoid as done in the congruent-arc technique of the Latarjet procedure.

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

We recommend the congruent-arc technique. Ultimately, the inferior aspect of the coracoid becomes part of the glenoid face. After the ideal position (between 3 and 5 o’clock on the glenoid) is achieved, a 2.5-mm drill is used to create two bicortical anteroposterior holes parallel to the glenoid articular surface. For a successful technique, Kocher clamps on either side of the coracoid (one on the glenoid side and another on the alternative side) should be used during drilling with the 2.5-mm drill. The use of these clamps allows for a trajectory that will not compromise the bone. Alternatively, commercial devices are available that hold the coracoid while you drill. Fixation is achieved with two 3.5-mm cortical or 4.0-mm malleolar screws, typically 34 to 36 mm in length, and washers with preloaded Suture Washers (Arthrex, Naples, FL) ( Fig. 1E.8 ). The screws should be snug but not overtightened. Any lateral overhang of the coracoid may be smoothed over with a high-speed bur.

Fig. 1E.8

A, Coracoid graft securely fixed to glenoid with two screws with no lateral overhang. B, Lateral view of the reconstruction.

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Capsular and Subscapularis Repair

A well-repaired capsule should allow the graft to function as an intraarticular platform and help protect the humeral head articular cartilage from the abrasive bone block ( ). Next, with the arm adducted to the side and placed in approximately 45 degrees of external rotation and the Fukada removed, the capsular repair is performed using the #2 FiberWires preloaded in the Suture Washers as well as with additional free high-strength #2 sutures to the capsule and the CA ligament. To allow for the imbrication of the capsular tissue, the sutures are placed in a figure-of-eight style. For further reinforcement, the CA ligament remnant on the coracoid graft is incorporated into the capsular repair. Finally, with the conjoined tendon exiting medially and anteriorly through the split, the subscapularis is repaired over the coracoid transfer with high-strength #2 suture. When repairing the lateral-most extent of the subscapularis split, care should be taken to avoid the long head of the biceps tendon. After copious irrigation, the wound is closed in a standard layered fashion.

Postoperative Management and Rehabilitation

The main goal of postoperative management and rehabilitation is to allow graft union. Because of the subscapularis split, a subscapularis protection protocol is usually not required. Moreover, a simple well-padded abduction sling is used to prevent strain, encourage rest, and reduce the risk of hematoma formation during the first 3 to 4 postoperative weeks. Gentle passive, active, and active-assisted shoulder ranges of motion in the scapular plane are permitted in this period, and motion of the fingers, hand, and elbow is encouraged. To reduce the risk of graft nonunion, no resisted elbow flexion is allowed for at least 6 weeks. Also, to optimize osseous healing, antiinflammatory pain medications are avoided in the early postoperative period. After serial radiographic evaluation confirms continued maintenance of graft placement, patients are allowed to resume progressive strengthening and conditioning. At the 4-month mark, return to contact sports and heavy labor exertion is allowed. We typically obtain postoperative imaging, including radiographs as well as two-dimensional and 3D CT scans, at 4 to 6 months after surgery to assess for final coracoid union ( Figs. 1E.9 and 1E.10 ).

Fig. 1E.9

Postoperative radiographs in the true anteroposterior ( A ), lateral ( B ), and Bernageau ( C ) views demonstrating optimal placement of the coracoid bone block with no lateral overhang and bicortical screw fixation. Radiographs from Gilles Walch.

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Fig. 1E.10

Example of axial view computed tomography (CT) ( A ), sagittal view CT ( B ), and three-dimensional CT ( C ) images at 6 months demonstrates a healed coracoid bone block (Latarjet).

From Gendre P, Thélu CE, D’Ollonne T, Trojani C, Gonzalez JF, Boileau P. Coracoid bone block fixation with cortical buttons: an alternative to screw fixation? Orthop Traumatol Surg Res . 2016;102(8):983-987.

Key Technical Pearls and Complications

Several steps have been proven to be crucial for efficacy and prevention of complications caused by technical error when performing the Latarjet procedure. First among these is performing a subscapularis split instead of a takedown or division. Full and partial takedowns of the subscapularis insertion require protection from passive external rotation for at least 6 weeks in addition to a graduated internal rotation strengthening program ( ). A split results in less morbidity while also having the added benefit of the sling effect ( ). Splitting is also done instead of dividing the subscapularis to avoid fatty infiltration of the subscapularis. Aside from less fatty degeneration, splitting has been shown to lead to significantly better Rowe and Walch-Duplay scores, which assess strength and functionality, respectively ( ). In particular, a subscapularis L-shaped tenotomy has been shown to lead to 28% less strength of the rotator cuff muscle at long-term follow-up compared with a subscapularis split ( ). Additionally, the decortication step is a key pearl that allows for improved graft union. Graft union is also promoted and graft lysis avoided by limiting pectoralis minor release to the tip of the coracoid process, thereby minimizing graft devascularization. Finally, extracapsular placement of the coracoid graft with suture anchors may theoretically soften the articulation contact pressures on the humerus, although this has not yet been clinically confirmed to result in a reduced incidence of degenerative arthritis.

Neurologic injury incidence after the Latarjet procedure has been reported to vary from 1.8% to 10.4% ( ). using intraoperative neuromonitoring during the Latarjet procedure found that the axillary and musculocutaneous nerves are at particular risk, especially during glenoid exposure and graft insertion. Given these risks, be sure to identify and release the musculocutaneous nerve from the posterior fascia of the conjoined tendon. Thus, neuromonitoring may have a role in the future of the procedure, particularly in cases with extensive scar tissue and anatomic distortion ( ).

Open versus Arthroscopic Latarjet

The evolution of modern arthroscopic techniques has led to their application in a completely arthroscopic Latarjet procedure with reported benefits such as decreased stiffness, quicker rehabilitation, and accelerated return-to-sport activities ( ). However, given the high complexity and required dexterity for this procedure, it is difficult to master and standardize ( ). Such is the case that a comparative and learning curve analysis of the open versus all-arthroscopic Latarjet procedure found that 10 arthroscopic Latarjet procedures were needed to overcome the need for conversion to the open version and 20 procedures were necessary before an equal operating time to the open technique was achieved ( ). Additionally, despite similar functional and patient satisfaction outcomes between both techniques, the arthroscopic cohort was measured as having higher incidence rates of complications, screw placement inaccuracy, persistent apprehension, and recurrences ( ). A systematic review and cost analysis of open versus arthroscopic Latarjet evaluating 1317 patients found similar good outcomes between the two, but direct costs of the arthroscopic procedure were found to be double that of the open operation on average ( ).

Glenoid Bone Loss: Arthroscopic Latarjet Procedure

Laurent Lafosse, MD
Johannes E. Plath, MD


The authors of this chapter review the indications, techniques, complications and post-operative rehabilitation of the arthroscopic Latarjet procedure for anterior shoulder instability. The authors describe a detailed guideline with eight defined steps in performing their technique.

Keywords: anteroinferior capsulolabral repair; arthroscopy; Bankart repair; bipolar bone loss; glenoid bone loss; labral repair; latarjet procedure.


  • The arthroscopic Latarjet procedure is the evolution of the classic open technique, providing all benefits of arthroscopic surgery. The first all-arthroscopic Latarjet procedure was performed in 2003. Since then technique and instruments have continuously been refined over the years, most recently in 2016. The current technique is performed in eight defined surgical steps via seven portals. A conversion to an open Latarjet is possible during every step throughout the procedure. The instrumentation allows both open and arthroscopic techniques.

  • The arthroscopic Latarjet procedure has shown excellent results through midterm follow-up, a low complication rate, and good graft positioning.

  • The arthroscopic Latarjet procedure is an advanced arthroscopic technique requiring extraarticular soft tissue dissection in close proximity to the brachial plexus. A detailed knowledge of the anatomy and a high level of experience in arthroscopic shoulder surgery are mandatory to successfully perform this procedure.


The typical lesion after an anterior shoulder dislocation is the Bankart lesion, a detachment of the capsulolabral complex from the anterior glenoid. Arthroscopic suture anchor repair reconstitutes the capsulolabral anatomy by reattachment of the anterior soft tissue to the glenoid. Excellent clinical outcomes can be expected. Over the past decade, the arthroscopic Bankart repair has become the standard treatment method of shoulder instability for many surgeons.

However, in many cases, the surgeon is faced with more extensive soft tissue damage such as complex labral tears with disruption of the labral ring; anterior labroligamentous periosteal sleeve avulsion (ALPSA) lesions;, a torn, attenuated, or avulsed inferior glenohumeral ligament from the humerus (humeral avulsion of the glenohumeral ligament [HAGL] lesion); or even a combination of multiple lesions. These high-grade soft tissue defects make an anatomic repair of the capsulolabral complex impossible.

Moreover, bone loss at the anterior glenoid after shoulder dislocation caused by fracture and erosion adds to the complexity of shoulder instability. A glenoid defect of 20% to 25% of the glenoid diameter is usually referred to as the critical defect size for shoulder stability ( ). However, isolated glenoid bone defects are rather the exception. Generally, we find additional humeral head impression fractures (Hill-Sachs lesion) that carry the risk of engagement and recurrence, especially if the “save arc” that the glenoid provides for humeral rotation is reduced (bipolar bone loss). The impact of glenohumeral bone loss on shoulder instability and failure of surgical treatment is well documented in the literature ( ).

In our experience, bone loss is rather the rule than the exception in patients with chronic shoulder instability and is commonly underestimated. investigated the incidence of glenoid rim lesions in 100 consecutive patients with recurrent anterior shoulder instability and found a glenoid defect in 90% of patients (50% bony Bankart lesion, 40% glenoid erosion).

In the complex situation of high-grade soft tissue and bony pathologies, a simple Bankart repair that reattaches the capsulolabral tissue to the glenoid cannot adequately address the complex underlying pathologies. Furthermore, a successful anatomical repair relies on good bone and soft tissue quality. However, tissue quality in patients with chronic shoulder instability is usually poor, and avulsed glenoid fragments are often subject to significant resorption and necrosis, which in principle prohibits an anatomic repair. The mentioned reasons may explain the significant number of persisting instabilities and recurrence in long-term follow-up studies of arthroscopic Bankart repair ( ).

Besides pathomorphologic findings, the patient’s risk factors for failure as well as the individual patient expectation of surgery need to be taken into account. Within the literature, several authors have reported inferior outcomes and increased rates of recurrence after arthroscopic Bankart procedures in young patients, male patients, patients with general ligamentous laxity, patients with a high number of preoperative dislocations, and patients with participation in contact and overhead activities ( ).

The Latarjet procedure involves the transfer of the coracoid with the attached conjoined tendon through a subscapularis split onto the anterior glenoid. In long-term studies, this procedure has proven its ability to reliably stabilize the shoulder joint, showing recurrent dislocation in fewer than 3% at 20 years of follow-up ( ). The success of the Latarjet procedure is a result of several factors. First, the vascularized coracoid bone graft reconstructs the deficient anterior glenoid rim and enlarges the glenoid articular arc and therefore prevents a large Hill-Sachs lesion from engagement. Second, the intersection of the transferred conjoined tendon and the inferior subscapularis reinforce the insufficient inferior capsule and forms a dynamic sling system that creates dynamic tension as the shoulder moves into external rotation and abduction. Biomechanical studies clearly demonstrate the importance of the soft tissue sling to fully restore shoulder stability, with more than 50% of stability being attributed to the sling effect in external rotation, mostly when the arm is in 90 degrees of abduction ( ).

With a combination of bony and soft tissue grafting, the Latarjet procedure addresses a broad spectrum of various capsulolabral and osseous pathologies around the shoulder and allows the treatment even of complex pathomorphologic findings. The arthroscopic Latarjet procedure is the natural evolution of the classically open procedure, providing the following advantages over the open Latarjet procedure ( ):

  • Accurate bone block placement under direct visualization from different perspectives

  • Treatment of concomitant lesions (e.g., superior labrum anterior and posterior [SLAP], posterior Bankart, rotator cuff)

  • General arthroscopic advantages because of the minimal invasiveness of surgery. These may include less postoperative pain, better cosmesis, less scarring and fewer deep adhesions, earlier mobility, fast rehabilitation and return to sports, and reduced risk for infection.

  • Decision making for the Bankart or Latarjet procedure during diagnostic arthroscopy depends on Hill-Sachs engagement, extent of capsulolabral lesion and soft tissue quality, which is difficult to predict on preoperative imaging.


Indications include ( ):

  • Significant glenoid bone loss (chronic Bankart fracture, glenoid erosion) as measured with three-dimensional (3D) computed tomography (CT)

  • Bipolar bone loss with Hill-Sachs engagement

  • Complex and irreparable capsulolabral lesions (labral ring disruption, destruction of labrum, ALPSA lesion, HAGL lesion, inferior glenohumeral ligament insufficiency)

  • Activity in overhead and contact sport with high risk of recurrence

  • Revision of failed soft tissue repair


  • Plain radiographs (true anteroposterior and lateral view, axial Bernageau view)

  • 3D CT to assess bony deficiency

Surgical Technique

Seven standardized portals are needed for the arthroscopic Latarjet procedure ( ) ( ):

  • A portal: Standard posterior portal at the soft spot of the posterior shoulder

  • D portal: anterolateral portal in line with the superior border of the subscapularis tendon

  • E portal: classic anterior instrument portal through the rotator interval

  • H portal: This portal is the “high” portal, superior to the coracoid, used for coracoid preparation and osteotomy.

  • I portal: Anterior viewing portal, “inferior” to the coracoid. The portal is located at the apex of the anterior axillary fold and established under direct vision with a spinal needle that is directed toward the coracoid.

  • J portal: Additional anterior viewing portal used during graft shaping as well as graft transfer and fixation. The portal is located on an arc midway between the D and I portals.

  • M portal: “Medial” to the conjoined tendon, this portal is located in line with the glenohumeral joint axis and midway between the inferior border of the pectoralis major and clavicle. The portal is used to insert the Double-barrel Cannula (DePuy Synthes Mitek, Raynham, MA).

  • Video 1F.1

    Arthroscopic Latarjet procedure using the Glenoid Guide (DePuy Synthes Mitek, Raynham, MA).

The arthroscopic Latarjet procedure consists of eight defined key surgical steps:

  • 1.

    Joint evaluation and confirmation of indication: The arthroscope is inserted in the A portal, and a thorough joint evaluation is performed. A probe may be introduced via the E portal. Rotator cuff tears, glenoid and humeral chondral defects, and labral and capsular lesions are noted. Particular attention is paid to the size and location of the Hill-Sachs lesions and anterior glenoid bone loss. Potential Hill-Sachs engagement is assessed with abduction and external rotation of the shoulder.

  • 2.

    Glenoid preparation and anterior exposure: The anteroinferior labrum is resected between the 2 and 5 o’clock positions. The capsule is detached from the glenoid neck, and the 5 o’clock position is clearly marked (α-mark) with the electrocautery. It is mandatory to do this step now with the camera in the A portal and facing toward the glenoid to have a view with minimal optical distortion. At the upper border of the subscapularis, the rotator interval is opened. The clavipectoralis fascia is incised along the lateral border of the conjoined tendon, opening the space between the coracoid process and pectoralis major. The scope is moved to the D portal. With the instruments in the E portal, labral resection and capsular detachment are finalized, and the glenoid is flattened for graft placement.

    After sufficient working space between the conjoined tendon and pectoralis major has been created, the anteroinferior portals (I, M, and J portals) are placed. For M-portal placement, it is important to stay anterior to the pectoralis minor to avoid violating the plexus. Furthermore, the surgeon should be aware of the glenohumeral joint line. To determine the orientation of the glenoid, a switching stick or probe can be introduced in the A portal. Alternatively, the assistant may place her or his index finger at the A portal to guide the right direction.

  • 3.

    Coracoid preparation: The scope is switched to the I portal. With the instrument in the M portal, the medial edge and, with the instrument in the J portal, the lateral edge of the coracoid can be accessed. For coracoid preparation, the arm rests in slight internal rotation. Medially, the interval between the conjoined tendon and pectoralis minor is dissected. Attention should be paid to avoid harming the musculocutaneous nerve. The interval is opened cautiously, and deep dissection is performed bluntly until the nerve has been located. Then the pectoralis minor tendon is detached. Laterally, the coracoacromial ligament is detached from the coracoid. Finally, all bursa and fat tissue is cleared from the superior and inferior aspects of the coracoid process, sparing the insertion site of the coracoclavicular ligaments. At this stage, the arm may be positioned in slight forward traction to expose the space lateral to the coracoid.

  • 4.

    Coracoid osteotomy: The arm is in internal rotation without traction. The H-portal is determined with a needle and incised large enough to accommodate the Inline Coracoid Drill Guide (DePuy Synthes Mitek). The Coracoid Drill Guide is inserted and placed flush on the coracoid in its long axis between the middle and medial third of the width of the coracoid, leaving adequately room to the coracoid tip. The 7-mm offset pin of the guide can be used to facilitate placement. Two 1.5-mm K-wires are inserted bicortically. The α (distal) and the β (proximal) holes are drilled bicortically using the coracoid step drill, the holes tapped, and a “top hat” washer placed into each hole over a K-wire ( Fig. 1F.1 ) Because there is always some resorption of the upper part of the graft, we like to countersink the superior β-screw in the graft. Accordingly, our preference is to only use a single washer at the α hole.

    Fig. 1F.1

    Coracoid preparation: view from the I portal. A “top-hat” washer has been placed in the α-coracoid hole. A cannulated drill is used to drill the β-coracoid hole. ct, Conjoined tendon.

    Using a curved sharp osteotome, stress risers are created at the medial and lateral cortex at the desired level of the osteotomy, anterior to the insertion site of the coracoclavicular ligaments. Finally, the osteotomy is completed between the two stress risers with the osteotome.

  • 5.

    Subscapularis split: Before the subscapularis split, it is mandatory to expose the axillary nerve that crosses the subscapularis at the level of its myotendinous junction as well as the superior and inferior border of the subscapularis tendon. The subscapularis is split via the M portal between the inferior and the medial third of the tendon, starting lateral to the axillary nerve. A probe can be introduced through the H or the A portal (or both) to elevate the upper part of the tendon and facilitate this split. Laterally, special caution is required not to harm the cartilage of the humeral head. A subscapularis channeler is introduced through the M portal and placed at the future graft location. To fully achieve the subscapularis split at the level of the muscle, the arm is brought to maximum external rotation, with attention paid to avoid dislocating the shoulder.

  • 6.

    Screw-hole placement for graft fixation: The screw holes can be placed using the Glenoid Guide (DePuy Synthes Mitek) or directly using the Double-barrel Cannula. For screw-hole placement, the table is tilted backward, which brings the scapular automatically in a position of scapular retraction. In addition, the shoulder is in internal rotation, and the elbow is in line with the scapula. This makes it easier to place the holes parallel to the joint axis. Two options are available to prepare the α hole:

    • a.

      Directly with the Double-barrel Cannula: The cannula is introduced into the M portal. A probe with a 5-mm tip is inserted through the A portal, aligned parallel to the glenoid and hooked onto the anterior glenoid at the α-mark. The cannula is placed medial to the probe tip, parallel to the glenoid ( Fig. 1F.2 ). The inferior coracoid screw is inserted into the cannula, and a K-wire is placed through the coracoid screw and at the level of the α-mark (accordingly 7 mm medial to the cartilage). When the guide is perfectly parallel to the inferior glenoid, the wire is introduced through the glenoid and through the skin of the posterior shoulder. A scaled tube is inserted from posterior over the K-wire. The α hole is drilled over the K-wire. After the second cortex is perforated by the drill, the tube is advanced anteriorly up to a stopper. The wire is removed while the tube remains in the glenoid and through the subscapularis split.

      Fig. 1F.2

      Glenoid tunnel placement: view from the I portal. A probe is inserted through the A portal and hooked onto the anterior glenoid at the α mark. The cannula is placed 5 mm medial to the cartilage (medial to the 5-mm probe tip), parallel to the glenoid (g). hh, Humeral head; ss, subscapularis.

    • b.

      Glenoid guide : The glenoid guide facilitates a parallel alignment of the graft screws to the glenoid and allows three offset options for graft placement (flush or 2 mm or 4 mm medialized to the articular surface). The guide is inserted into the M portal. A K-wire is inserted through the offset pin at the α-mark, advanced into the glenohumeral joint, and aligned parallel to the joint line. A second K-wire is placed in the appropriate cannula (0, 2, or 4) hole medially. As previously described, the K-wire is inserted into the glenoid and overdrilled, and a scaled tube is inserted posteriorly.

    Whatever technique is used, it is mandatory to control the correct position (vertical and horizontal introduction point and orientation) of the K-wire before drilling the hole, and the surgeons should not hesitate to repeat the wire placement to ensure accuracy.

  • 7.

    Coracoid placement: The scope is moved to the J portal. With help of two 3.5-mm coracoid screws inserted through the cannula’s α and β holes, the coracoid graft is firmly fixed to the cannula, but the tip of the cannulated coracoid screws should not extend past the far cortex. The coracoid is mobilized to ensure that there are no persisting soft tissue restrictions or fascial bands. A K-wire is inserted from anteriorly through the α-coracoid screw and inserted in the scaled glenoid tube. This connection between the cannula and glenoid stabilizes the graft during graft shaping. The coracoid graft that will meet the glenoid neck is now smoothened and shaped with a bur.

    The cannula is turned 90 degrees to orient the coracoid graft horizontally, and smoothly advanced through the subscapularis split along the K-wire pushing the scaled glenoid tube posteriorly. The graft is placed parallel and flush to the anterior glenoid. If necessary, the coracoid is removed back to the split and reshaped. When the coracoid fits perfectly, a second K-wire is drilled through the glenoid at the level of the β-coracoid screw to additionally fix the graft. It should be parallel to the first wire.

  • 8.

    Coracoid fixation: The α-coracoid screw is unscrewed and removed, and the scaled tube is again advanced anteriorly up to the bone stopper. The required α screw length can be read directly off the scale. A screw of the right length is inserted, and the graft is fixed inferiorly. In the same way, the β hole is drilled and measured, and the β screw placed superiorly.

Finally, the graft and screw position is evaluated. If the graft is too proud, it can be smoothened with the bur ( Fig. 1F.3 ).

Fig. 1F.3

Final result: view from the I portal. The graft has been flattened to the level of the subchondral cartilage of the glenoid (g). The probe is left in the posterior portal to elevate the upper subscapularis (ss). cg, Coracoid graft; hh, humeral head; S, screw.

Pearls and Pitfalls

  • To improve working space: Place a switching stick in D portal to elevate the deltoid muscle and add slight forward flexion and traction during anterior dissection.

  • To decrease fluid loss and improve visualization: Seal the portals with plastic stoppers or cover the portals with a finger.

  • The arthroscopic Latarjet is “teamwork”: Perfect collaboration between the anesthesiologist and the surgical team is mandatory to obtain perfect visualization (e.g., blood pressure vs pump pressure management).

  • Partial superior graft osteolysis is a common finding after a Latarjet procedure ( ). In our experience, this does not affect shoulder stability; however, the superior screw may get in contact with the subscapularis tendon. In cases with minor glenoid bone loss and expected partial graft resorption, respectively, we do not use a “top-hat” washer for the superior screw. This allows us to countersink the screw into the graft, so the head of the screw is less proud in case of resorption.

  • Perform 6-month follow-up CT imaging and remove screws if necessary.

  • If in doubt, you can always convert to an open Latarjet procedure. It is much better to convert and to manage a good open Latarjet procedure rather than performing a bad arthroscopic Latarjet procedure!

Rehabilitation and Follow-Up

  • Return to daily activities and work immediately after surgery depending on postoperative pain; no restrictions of mobility

  • Return to low-impact sports after 6 weeks to 3 months

  • High-risk activities, including contact and overhead sports, at 4 to 6 months


The senior author and innovator of the technique has performed more than 700 cases of the arthroscopic Latarjet procedure to date. The longest systematically followed-up series (minimum follow-up, 5 years) was published in 2014, providing outcomes of 64 of the 89 shoulders operated before June 2008 ( ), representing early results during evolution of the technique and the innovator’s own learning curve.

At a mean follow-up period of 6.4 years after surgery, the average Western Ontario Shoulder Instability Index (WOSII) was 90.6% ± 9.4%. One patient (1.6%) complained of persisting subluxations. No patient had a recurrent dislocation. Eight shoulders (12.5%) needed screw removal because of partial graft resorption, and one patient needed revision because of secondary graft dislocation. One patient underwent total shoulder replacement for dislocation arthropathy during follow-up. There were no neurovascular complications.

Glenoid Bone Loss: Bone Graft Options: Iliac Crest and Distal Tibial Allograft

Matthew T. Provencher, MD
Anthony Sanchez, BS
George Sanchez, BS
Katrina Schantz, PA-C


Although coracoid transfer techniques, namely the Bristow and Latarjet procedures, are considered the first-line procedures of choice for treatment of anterior shoulder instability with significant bone loss, these procedures have been shown to fail as a result of graft resorption, graft malpositioning, and fixation screw loosening. In the case of a previous failed coracoid transfer procedure or if the glenoid osseous deficiency is too large to be treated via the Latarjet procedure, alternative bony augmentation procedures may be considered. Here we describe two glenoid bony reconstruction techniques that can be completed as alternatives or revisions to coracoid transfer: (1) anatomic glenoid reconstruction via fresh osteochondral distal tibial allograft and (2) iliac crest bone autograft. A comprehensive review of the indications, workup, surgical technique, pearls, and potential complications associated with each of these two techniques is provided.

Keywords: bone graft; distal tibia; failed Latarjet; glenoid bone loss; iliac crest; recurrent shoulder instability.


  • Many argue that coracoid transfer techniques, namely the Bristow and Latarjet procedures, poorly reconstitute the glenoid arc and chondral surface; others claim these procedures significantly alter the native anatomy because of the placement of the conjoined tendon and the extraarticular capsulolabral repair.

  • Although coracoid transfer techniques have been proven effective for treatment of glenoid bone loss, these procedures do fail as a result of graft resorption, graft malpositioning, and fixation screw loosening.

  • Alternative bone graft augmentation procedures that may be implemented in the case of a failed Latarjet procedure include (1) anatomic glenoid reconstruction via fresh osteochondral distal tibial allograft (DTA) and (2) iliac crest bone autograft (ICBG).

  • The indications, workup, surgical technique, pearls, and potential complications associated with each of these two bony augmentation techniques for treatment of recurrent shoulder instability with bone loss are reviewed in this chapter.

Indications and Preoperative Workup

The main indications for bone grafting using DTA or ICBG include a history of failed Latarjet procedure, an osseous deficiency, or fracture that involves more than one third of the glenoid. Post–Latarjet procedure breakage of fixation screws, coracoid graft resorption, and graft nonunion all warrant revision surgery featuring one of these two grafts ( ). It is difficult to state when these are the best options. Randomized trials are needed to expand the pool of patients that can be offered these techniques as a primary treatment. Early work described by Provencher et al (2001b) suggest that DTA may be a viable treatment option for postsurgical Latarjet failure.

The preoperative workup necessary for an anatomic glenoid bone reconstruction using DTA or ICBG closely resembles that of coracoid transfer techniques. Physical examination should evaluate anterior stability and ruling out of other possible pathologies contributing to the presenting sequelae. Imaging should include three-dimensional computed tomography (3D CT), imaging studies featuring the glenoid fossa circle method to quantify glenoid bone loss ( ), and arthroscopic bare spot landmark diameter-based measurement ( ). All contribute to a comprehensive surgical plan before performing a glenoid bone reconstruction using DTA or ICBG.

The DTA offers multiple advantages as compared with ICBG. Along with the inferior aspect of the coracoid, the lateral DTA has been shown by cadaveric studies to have the best-match radius of curvature to reestablish the native glenoid ( ). More clinical evidence is needed to confirm its validity with respect to long-term outcomes, but controlled laboratory studies have also shown that DTA exhibits improved joint congruity and lower peak forces within the glenohumeral joint compared with Latarjet reconstruction. Postoperative evaluation of patients with severe glenoid bone loss treated through DTA demonstrates positive outcomes, including significant improvements in American Shoulder and Elbow Society score, Western Ontario shoulder instability index, and single numerical assessment evaluation score (<0.01) at an average follow-up period of 45 months. Moreover, the healing rate of the allograft to the native glenoid was 89% with no cases of recurrent instability ( ). Tibial allografts are more readily available than fresh glenoid allografts because of greater graft contamination issues with tissues close to the gastrointestinal tract and lungs ( ). The distal tibia has weight-bearing bone with dense trabecular matrix, making it more accommodating for drilling while having the potential for improved bony incorporation ( ). In addition to the bony advantages, robust weight-bearing cartilage allows for a cartilaginous interface between the humerus and glenoid allograft, making the DTA a potentially ideal glenoid resurfacing graft ( ).

Evidence that can inform the decision when considering DTA versus inner table ICBG was found via cadaveric biomechanical studies in the context of posterior glenoid reconstruction. Similar contact mechanics were measured among the two ( ). Cadaveric biomechanical studies in the context of posterior glenoid reconstruction provide similar evidence when comparing DTA and ICBG.

Patient Positioning and Surgical Approach

After anesthesia is induced (with an interscalene regional block if possible), the patient is placed in the beach-chair position with approximately 40 degrees of head elevation. Two towels are placed between the patient and the bed behind the medial border of the scapula to flatten, and more important, stabilize the glenoid and scapula, preventing unwanted anterior and internal rotation. Restricting anterior rotation of the glenoid ensures an optimal trajectory for working on the anterior aspect and allows for screw fixation of the graft with ease. The arm may be placed in a commercially available specialized arm holder or left free and controlled on either a padded Mayo stand or an adjustable operating room tray. A typical open Bankart incision of approximately 7 cm is made with a #10 scalpel, beginning from the tip of the coracoid process and extending directly inferiorly to the superior axillary fold.

After identifying the deltopectoral interval, the cephalic vein is preserved and gently retracted laterally. The fascia overlying the conjoined tendon is identified and incised with Metzenbaum scissors. Gelpi retractors are placed to expose the fascia, while mobilizing the subfascial plane of the deltoid. Next, the lateral aspect of the conjoined tendon is identified, incised just lateral to the muscle belly of the short head of the biceps, and retracted medially. The deltoid is retracted laterally with a Kolbel retractor placed between the deltoid and the conjoined tendon. Care should be taken to protect the musculocutaneous nerve by avoiding excessive medial retraction. The subscapularis is then identified and split in at the junction of the superior two thirds and middle third longitudinally in line with its fibers. A subscapularis takedown may be necessary in the revision setting, but subscapularis takedown or tenotomy should be avoided. When using the subscapularis splitting approach, caution should be taken to avoid splitting medial to the coracoid due to potential iatrogenic nerve injury (nerve supply to subscapularis). A subscapularis splitting approach should provide sufficient exposure to the glenoid neck. However, if needed, the superior two thirds and even the entire subscapularis may be taken off the humerus. A pointed retractor is placed in the split to gain access to the capsule, which is freed from the posterior aspect of the subscapularis. The glenohumeral joint capsule is more easily separated from the subscapularis medially.

After separating the capsule from the subscapularis medially, it is also separated laterally, which becomes difficult because of the close adherence of the subscapularis to the capsule laterally. A medial-based T-shaped capsulotomy is performed for exposure. To facilitate this, internal rotation, as well as gentle lateral traction may be used. Using a #15 scalpel, the capsule is elevated off the glenoid neck medially in a subperiosteal fashion. Subsequently, care is taken to protect the axillary nerve, the labral tissue is dissected and elevated medially because it will be repaired to the anterior aspect of the bone graft after fixation. In revision cases, anterior labral tissue may be deficient, not allowing repair. If in a Latarjet revision setting, hardware is now removed. An elevator and rongeur are used to debride the scar tissue and prior implants. A high-speed bur and rasp are used to establish a uniform, bleeding surface on the anterior glenoid rim as perpendicular as possible to the glenoid articular surface. Having identified the anterior glenoid, glenoid bone loss seen on the preoperative 3D CT scan is confirmed. Generally, a 25% to 30% loss indicates the need for approximately 8 to 9 mm of anterior glenoid bone augmentation ( ). Using the central bare spot as the landmark for assessing bone loss of the inferior glenoid provides custom dimensions of the allograft that is to be prepared. In addition, sizing blocks and preoperative measures are used to confirm the bone block size ( ).

Distal Tibial Allograft Preparation

A fresh distal tibia is obtained from a donor cadaveric source ( ). It is not necessary to size-match the DTA because it has a radius of curvature and sizing that has been shown to work well in the vast majority of cases ( ). Using the confirmed preoperative 3D CT scan and intraoperative bare spot measurements, the lateral third of the distal tibia is fashioned into a bone graft. After the distal tibia is marked ( Fig. 1G.1 ), a 0.5-inch sagittal saw is used to cut while an assistant holds the graft in place with two towel clamps. While cutting, the allograft must be kept cool with irrigation. The allograft may be cut at various angles to accommodate any slope changes of the glenoid, but is usually 5 to 15 degrees angled. Typically, between 8 and 11 mm of tibial bone is used. Next, the superior- and inferior-most aspects of the allograft are rounded to the shape of the glenoid. Following pulsatile lavage, two 4.0-mm drill holes are drilled in the graft while the graft holder is used as a guide. To facilitate placement and positioning within the joint, two 1.6-mm K-wires are placed in the allograft bone at a 45-degree angle to the articular surface. The allograft is then placed in the position and orientation in which it will be fixed onto the native glenoid with a final assessment of fit, conformity, and angle relative to the articular surface.

Fig. 1G.1

A, A surgical marker is used to mark a 10-mm measurement of the lateral chondral surface. B, An angular cut is marked at 5 to 10 degrees toward the medial tibia to retrovert the graft. C, Lateral view of the graft.

From Cole BJ, Sekiya JK: Surgical Techniques of the Shoulder, Elbow, and Knee in Sports Medicine . Philadelphia: Saunders; 2013.

Glenoid Preparation And Graft Fixation

After any necessary adjustments, two small 1.6-mm K-wires are placed to temporarily hold the graft away from the future locations of screw placement ( Fig. 1G.2 ). The graft is placed flush with the glenoid. The K-wires are measured to select the appropriate screw length, typically 32 to 38 mm. A cannulated drill is used over the K-wires, through the 4.0-mm holes drilled earlier in the graft, onto the glenoid. The graft is then fixed via a lag technique with two 3.75-mm noncannulated fully threaded cortical screws with Suture Washers (Arthrex, Naples, FL.) ( Fig. 1G.3 ). After fixation, if the capsule and labrum are available for repair, they are repaired using the sutures from the Suture Washers. The subscapularis is then closed with multiple suture anchors. In our experience, a drain is not necessary if the deep wound remains dry. Pearls and pitfalls for this procedure are reviewed in Table 1G.1 .

Fig. 1G.2

Kirschner wires are placed at a 45-degree angle to facilitate placement of the graft.

From Cole BJ, Sekiya JK: Surgical Techniques of the Shoulder, Elbow, and Knee in Sports Medicine . Philadelphia: Saunders; 2013.

Fig. 1G.3

Intraoperative photograph of the final anatomic glenoid bone reconstruction using distal tibial allograft.

From Cole BJ, Sekiya JK: Surgical Techniques of the Shoulder, Elbow, and Knee in Sports Medicine . Philadelphia: Saunders; 2013.

Table 1G.1

Pearls and Pitfalls for a Successful Anatomic Glenoid Bone Reconstruction via Distal Tibial Allograft

Pearls Pitfalls
Size-matching DTA using preoperative 3D CT and intraoperative bare spot measurements for optimal allograft size and dimensions Radius of curvature mismatch or the dimensions of graft will not restore anatomy and biomechanics if measurement steps not taken
Limiting exposure via subscapularis split versus take-down or division when possible Iatrogenic nerve damage if care is not taken
Protect the musculocutaneous nerve by avoiding excessive medial retraction.
Protect the axillary nerve when dissecting and elevating labrum.
Gentle neurolysis of the axillary nerve can be performed.
Intraoperative neuromonitoring to ensure no iatrogenic injury is caused
Irrigation and pulsatile lavage ensure minimal graft damage and clear marrow-free tunnels ready for placement.
Before graft fixation, platelet-rich plasma can be used to improve the consolidation process.

CT, Computed tomography; DTA, distal tibial allograft; 3D, three dimensional.

Postoperative Rehabilitation

A padded abduction sling is to remain on at all times for the first 4 to 6 weeks postoperatively. Rehabilitation consists of six phases. The first phase is 2 weeks in duration and characterized by prohibition of biceps activation and aerobic exercises, including stationary bike and level surface walking capped by 30 minutes. In this first phase, 4 weekly passive range of motion routines are performed up to specific limits, as follows: 120 degrees of forward flexion, 120 degrees scapular plane, 30 degrees external rotation at side, and 90 degrees abduction. Active wrist and elbow range of motion are encouraged in phase 1.

In phase 2, typically weeks 2 to 4, the aerobic exercise limits are progressed to 45 to 60 minutes. Goals are set to improve passive range of motion: 150 degrees of forward flexion, 150 degrees scapular plane, 45 degrees of external rotation at the side, and 90 degrees of abduction. Isometric exercises are recommended for extension, abduction, and external and internal rotation. Deltoid isometric exercises are allowed at the 4-week mark.

The third phase, weeks 6 to 12, commences with discontinuing the ultra-sling. This period increases the passive benchmarks of forward flexion, scapular plane, external rotation at side, and abduction to 160, 160, 45, and 140 degrees, respectively.

Confirming successful graft incorporation using serial imaging ( Fig. 1G.4 ), phase 4 adds an inclined treadmill to the rehabilitation plan. Passive range of motion exercises are transitioned to active-assisted and eventually, active range of motion.

Fig. 1G.4

Two- ( A ) and three-dimensional ( B ) computed tomography images illustrating an excellent match of the radius of curvature between the distal tibial allograft and native glenoid.

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

From weeks 12 to 16, phase 5, strength training begins. The regimen consists of push-ups, plyometric exercise, and external and internal rotation at 90 degrees with a cable. Having met the goals set by the previous phase, the sixth phase begins typically after 16 weeks. Swimming, military press, and pull-downs are permitted. Finally, throwing is allowed and gradually increased in distance.

Iliac Crest Autograft

Patient Positioning

After anesthesia is induced (with a regional interscalene block if possible), the patient is positioned on a full-length beanbag with the head of the bed elevated to 30 degrees. Using the beanbag as support, the aim is to ensure mobility of the shoulder and exposure of the ipsilateral iliac crest. Degree and direction of instability is confirmed via reexamination under anesthesia. The shoulder, arm, and ipsilateral iliac crest are prepared and draped in the standard sterile orthopedic fashion.

Surgical Approach

The ICBG surgical approach is the same as that of the DTA procedure.

Iliac Crest Exposure and Autograft Preparation

A curvilinear incision posterior to the anterosuperior iliac spine is made. Electrocautery is then used to gain access to the fascial plane where the abdominal oblique and the tensor fascia lata muscles join. Care should be taken to leave the abductor insertion intact. After the superior iliac crest is exposed and blunt retractors are used to maintain exposure, a tricortical wedge-shaped graft is removed using an oscillating saw and osteotomes. This wedge-shaped graft should measure approximately 3 cm long by 2 cm wide ( Fig. 1G.5 ). Using a small saw and bur, the graft is contoured in such a fashion that the concave inner table recreates the articulation. After harvest, the abdominal oblique muscle and tensor fascia lata are reattached followed by a layered closure, and finally the skin is approximated and closed with a running subcuticular stitch. To approximate the correct anatomic joint concavity, the graft-to-scapula interface angle is inclined. Care should be taken to not position the graft too acutely or horizontally to avoid possible impingement of the humeral head and failure to restore the glenoid concave depth, respectively. The inner table of the iliac crest is used as the glenoid face.

Fig. 1G.5

Illustration demonstrating harvest of the iliac crest autograft, approximately 3 cm long by 2 cm wide, from the greater pelvis.

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Glenoid Preparation and Graft Fixation

After adjustments to ensure anatomic approximation, two terminally threaded stainless steel 4.0-mm cannulated screw set K-wires with Suture Washers are used for fixation. To have the screw heads sit as far away medially from the humeral articular surface as possible, the wires are oriented medially and parallel to the joint surface. The transition between the autograft and native glenoid should be smooth visually and via palpation. A small bur is used to contour the transition if necessary. When the fixed graft is flush to the remaining glenoid ( Fig. 1G.6 ), the humeral head retractor is removed and the humeral head is positioned on the newly reconstructed glenoid. Joint congruity and stability are confirmed by rotating the arm. Next, using sutures from the Suture Washers, as well as horizontal mattress sutures, the capsule-periosteal sleeve is repaired to the edge of the graft. Chronic bone loss can cause shortening of the capsule, limiting lateral capsule repair. If capsule reattachment to the humeral neck with at least 30 degrees of external rotation is not possible, the lateral subscapularis is used to extend the capsule ( ). Using a #2 braided nonabsorbable horizontal mattress suture, the capsule is sewn into the undersurface of the lateral subscapularis so that the capsule becomes tensioned in external rotation ( ). Through use of 2 braided nonabsorbable horizontal mattress sutures and direct fixation to bone with Suture Anchors, the subscapularis split is repaired. Incision closure is completed in layers with subcuticular resorbable 3-0 monofilament suture and Steri-Strips. The incision is then dressed, and the shoulder is placed into an immobilizer. Pearls and pitfalls for this procedure are reviewed in Table 1G.2 .

Fig. 1G.6

Intraoperative photograph of final anatomic glenoid bone reconstruction using iliac crest autograft.

From Provencher MT, Romeo AA: Shoulder Instability: A Comprehensive Approach . Philadelphia: Saunders; 2012.

Table 1G.2

Pearls and Pitfalls for a Successful Anatomic Glenoid Bone Reconstruction via Iliac Crest Autograft

Pearl Pitfalls
Preoperative and intraoperative measurements inform the fashioning of the graft Nonanatomic glenoid bony reconstruction
Awareness and care of the neurovascular structures is necessary to prevent any possible damage. With neuromonitoring in extensive scar tissue revision setting when possible. Iatrogenic neurovascular injury as a result of lack of awareness. This especially holds true in cases of revision surgery after failed Latarjet procedures.
Leave the abductor insertion intact.
Conservative subscapularis management when possible.
Do not position the graft with too acute so as to not risk impingement of the humeral head.
Do not position the graft in a horizontally inclined interface angle to avoid risk of failing to restore glenoid concave depth.

Postoperative Rehabilitation

A shoulder immobilizer is used for the first 4 postoperative weeks with pendulum exercises permitted after the first week. Supervised physical therapy to regain range of motion begins after week 4, along with permitted arm use in routine daily tasks. Gradually, with serial imaging confirmation of graft healing, active and passive assisted motion are added to therapy. Water therapy is also encouraged after at least 6 weeks. Strength training is allowed after 3 months. Noncontact overhead recreational sports are permitted after 4 months. After at least 8 months, collision and contact sports are allowed.


Potential complications associated with these two bony reconstruction procedures discussed are similar to those of coracoid transfer procedures. These include graft nonunion, resorption of the graft, loosening or breakage of the screws, neurovascular injury, and injury to the articular cartilage. Additional concerns include enlarged or torn capsules and latent residual lesions that were not treated at the time of graft procedure.

Glenoid Bone Loss: Bone Graft Options Distal Clavicle

John M. Tokish, MD
Adam Kwapisz, MD, PhD
W. Stephen Choate, MD


The presence of bone loss is reported in 72% of instability cases, and many studies have reported that significant bone loss results in a higher failure rate than ∗∗∗. However, the amount of glenoid bone loss that should be considered as significant is still to be debated. Traditionally, glenoid deficiency of 20% was considered as well tolerated, but most recent data suggest that subcritical loss of 13.5% has a negative influence on clinical outcomes. Recently, the “on-track–off-track” concept has evolved as a result of glenoid bone loss and Hill-Sachs lesion interaction study. This concept, based on the fact that bipolar lesions significantly reduce shoulder stability when acting together, is clinically validated and has proved to be highly predictive of instability treatment outcome. There are numerous options reported to treat glenoid bone loss. The Latarjet and Bristow coracoid transfers, iliac crest bone grafting, and osteochondral allografts are the most described ones. Each has unique advantages, but none of them is an ideal option. An ideal graft should be readily available, free, without donor site morbidity, and able to resurface glenoid cartilage. This chapter describes the distal clavicular osteochondral autograft (DCA) as an alternative to the above-mentioned techniques. The graft has the ability to restore both the native glenoid radius and comparable amount of its cartilage. Because it is an autograft, this technique is a readily available method. It also proves favorable to the coracoid graft amount of corticocancellous buttress for glenoid arc restoration. Although DCA provides anatomic, theoretical, and biomechanical promises, longer term clinical studies are necessary to validate its use in the clinical setting. However, DCA has been used by the senior author in the setting of both anterior and posterior glenoid bone losses with a promising short-term outcome.

Keywords: autograft; distal clavicle.


  • Bone loss has emerged as a critical issue in the treatment of glenohumeral instability ( ). Its presence is reported in 72% of instability cases and influences the outcome of arthroscopic treatment ( ).

  • Many studies have reported that the presence of significant bone loss results in higher failure rates and lower clinical outcomes than patients without bone loss ( ).

  • The amount of glenoid bone loss that places an arthroscopic repair at risk has continued to evolve. Knowing that the glenoid diameter is 24 to 26 mm on average, a bone defect as small as 6 to 8 mm qualitatively translates into the “inverted pear” concept of Lo and Burkhart and is considered unacceptable by many ( ).

  • Traditionally, bone loss of up to 20% has been well tolerated, but greater amounts have been shown to affect biomechanical stability and clinical results ( ).

More recently, less bone loss, “subcritical bone loss” of 13.5%, has been shown to be detrimental to clinical outcomes ( ). It has also been shown in cadaveric study that less bone loss can compromise soft tissue repair ( ).

Hill-Sachs Lesion

  • Glenoid bone loss may also interact with the Hill-Sachs lesion of the humerus. Larger Hill-Sachs lesions may “engage” in functional positions and have also been shown to negatively affect arthroscopic results.

  • Recently, these two bone loss conditions have been combined into an “on-track–off-track” concept ( ). This paradigm takes into account bipolar bone loss, recognizing that the individual lesions can affect one another, reducing the stability of the shoulder joint.

  • This concept has been clinically validated, where applying the track concept was demonstrated to be highly predictive of outcome and more predictive than glenoid bone loss alone ( ).

  • Furthermore, patients undergoing an arthroscopic repair for an off-track lesion demonstrated Western Ontario Shoulder Instability Index scores 450 points worse than those who were on track at the time of surgery.

  • Thus, the recognition of the presence of bone loss is a critical aspect to the decision-making algorithm in the treatment of anterior shoulder instability ( ).

Measurement of Bone Loss


  • Glenoid and humeral bone loss may be evaluated both pre- and intraoperatively. Although special x-ray evaluations have been shown to characterize both glenoid bone loss and Hill-Sachs lesions, three-dimensional computed tomography (3D CT) scans have become the gold standard ( ).

  • Some authors suggest that magnetic resonance imaging (MRI) may be used to evaluate glenoid bone loss and that this evaluation can be as accurate as traditional methods ( ).

Technical Notes

  • Several methods of glenoid bone loss quantification have been proposed. described a method based on the assumption that after humeral subtraction, the inferior two thirds of the glenoid approximates a perfect circle. The authors then calculated the lost area of this circle as bone loss.

  • calculated the distance from the glenoid diameter to the defect (d) and evaluated bone loss as a function of d/radius.

  • Gerber measured the length of bone lesion (x) and the radius (r) of the glenoid and described that if x > r, there was less than 70% resistance to dislocation ( ).

  • Lo and Burkhart described an arthroscopic technique for measurement of bone loss based on the observation that the glenoid “bare area” is usually at its center. The glenoid bone loss may be calculated as the percentage of twice the intact radius ( ).

  • contradicted this, noting that the bare area is often not present, and when present, is rarely in the exact center of the glenoid. These authors recommended against using the arthroscopic bare area to calculate bone loss because it led to significant errors in measurement.

“On-Track–off-Track” Concept

  • recently described an on-track versus off-track concept of bipolar bone loss. In this setting, the authors compare glenoid track with the Hill-Sachs defect size. They note that when the Hill-Sachs lesion is larger than the glenoid track, the lesion is “off track” and is likely to engage.

  • demonstrated that off-track patients sustained a 75% failure rate with arthroscopic stabilization compared with an 8% failure rate in on-track lesions.

Surgical Treatment Options

  • There are number of options described to treat bone loss in shoulder instability. These include coracoid transfers such as the Latarjet and Bristow procedures, iliac crest bone grafting, and osteochondral allografts.

  • Each approach has advantages and disadvantages unique to its application. Factors such as size of graft, whether it has an inherent cartilage source, immunocompatibility, availability, and cost are all considerations in decision making for graft selection, and are discussed later ( Table 1H.1 ).

    Table 1H.1

    Comparison of Graft Types for Treatment of Glenoid Bone Loss in Shoulder Instability

    Distal Clavicular Autograft Coracoid Transfer Iliac Crest Transfer Allograft
    Cartilage source Patient’s own None None Allograft only; tibial or glenoid
    Versatility Anterior or posterior Anterior only Anterior or posterior Anterior or posterior
    Graft rejection None None None Potentially
    Cost Free Free Free Expensive give number if you can find a source
    Availability Readily Readily Readily 6–9 mo
    Sling effect None Conjoint tendons, subscapularis None None

  • The ideal graft should be such that it can restore the bone and cartilage loss seen in erosive glenoid bone loss with a source that is readily available, free, and without donor site morbidity.

Coracoid Bone Autograft

Historic Background

  • In 1954, Latarjet described his technique of coracoid bone transfer as a treatment of recurrent dislocation of the shoulder ( ). In 1958, Helfet described a coracoid tip transfer, and named it the Bristow technique ( ).

  • Although it has been more than half a century, coracoid transfer is still considered the gold-standard technique of glenoid bone loss treatment.


  • Coracoid grafts have the advantage of restoring a significant amount of glenoid bone and provide what has become known as the “triple blocking effect” of bone restoration, capsular reconstruction, and the sling effect of myodesing the inferior subscapularis with the conjoined tendon.

  • Biomechanical studies demonstrate that this procedure is quite effective in restoring shoulder stability, and clinical outcome studies have demonstrated low recurrence rates and excellent patient reported outcomes ( ).


  • The coracoid transfer is a nonanatomic approach, which may make future revision surgery more difficult, and its overall complication rate may reach 30% ( ).

  • There is a limit to the amount of glenoid surface that can be restored. reported that this technique may be insufficient to restore defects exceeding 31%. described that in comparison to the traditional Latarjet technique, its modification, the congruent arc technique, may be used to reconstruct bone loss of an approximate size of 54%, but this modification results in a greater graft displacement and a lower clinical failure load.

  • Other authors have shown that up to 60% of the graft may undergo osteolysis ( ).

  • Coracoid transfer lacks articular cartilage on the transferred graft. This has been cited as a potential reason for osteoarthritis development after Latarjet surgery, which is reported in up to 62% of cases ( ).

  • Unplanned reoperations have been reported at a higher rate than that of the traditional Bankart procedure, and the overall complication rate has been reported as high as 25% ( ).

  • have reported that neuromonitoring during Latarjet surgery resulted in a nerve alert in 77% of patients and resulted in 21% of patients with a clinically detectable nerve deficit postoperatively.

Iliac Crest Bone Autograft

Historic Background

  • The first reports of using bone block to restore glenoid deficiency were made by and .

  • In 2006, Warner et al reported autogenous tricortical iliac crest bone graft to be effective in treatment of recurrent instability in settings of glenoid bone loss.


  • This graft could restore defects up to 35 mm in length, which is significantly more surface than that of coracoid graft.

  • There are excellent short-term results with a few complications in 4% of patients. The technique has the advantage of being readily available, essentially free, and is an autograft source of bone ( ).


  • The iliac crest is nonarticular and thus cannot restore the osteoarticular loss seen in the glenoid. This may lead to secondary osteoarthritis, which has been reported after this procedure ( ).

  • There is the potential for donor site morbidity, from which persistent pain for longer than 1 year after surgery is described in up to 100% of cases.

  • Local infection (14%) and anterior superior spine fracture (3%) incidences have been reported ( ).

  • The ICBG graft does not, in itself, address the soft tissue pathology that is frequently present in instability cases where it is used ( ).

Distal Tibia Allograft

Historic Background

  • Distal tibial allografts have recently been introduced as an osteochondral source for glenoid bone loss treatment.


  • It may provide equivalent biomechanical properties to the iliac crest bone graft, and the technique has been shown to produce a better articular pressure profile than the Latarjet.

  • Glenoid arc articular conformity can be reproduced with this graft source ( ). Promising clinical outcomes have been presented but not yet published in the peer-reviewed literature.


  • The distal tibia may be less congruous than originally reported.

  • reported that the chance of a random pairing of a distal tibial allograft matching the radius of curvature of a recipient glenoid was low. How precise a match is necessary to achieve optimal results remains to be studied.

  • The possibility of graft resorption as well as immunologic response after this technique has not been investigated, but these concerns have plagued allograft usage in other transplant settings ( ).

  • The largest limitation to this method is its logistic application. The cost of a fresh osteochondral allograft can exceed tens of thousands of dollars, and there can be a significant wait time, which can exceed 6 months.

  • Fresh allograft preparation requires a minimum of 14 days as quarantine for infectious disease, and chondrocyte viability has been shown to significantly drop after 28 days postmortem. This requires surgeons to perform transplantation in roughly a 2-week window, which can be a scheduling challenge for both the patient and surgeon in many facilities ( ).

Distal Clavicle Osteochondral Autograft

Historic Background

  • The distal clavicle provides a fresh, osteochondral autograft in the treatment of glenoid bone loss ( ) ( ).

  • Video 1H.1

    Surgical technique for arthroscopic distal clavicle osteochondral autograft (DCA) implantation in the treatment of glenoid bone loss. The patient is positioned in the lateral decubitus position. This is a left shoulder using an anterior viewing portal through the rotator interval. After a standard diagnostic evaluation of the shoulder is performed, the decision is made to perform a DCA implantation. A glenoid neck is biologically prepared to create healthy bed of flat bleeding cancellous bone. After it has been created, an osteochondral autograft is harvested from the patient’s ipsilateral clavicle according to the Mumford technique. Two anchors are implanted so to secure a graft at the glenoid neck and DCA is being delivered with a “double-pulley” technique. When it is done, it is secured by tying the remaining opposite suture limbs. Portals are then closed arthroscopically and instruments removed.


  • The distal clavicular autograft (DCA) is the first reported option that provides an autograft source of bone and cartilage to replace similar tissue loss on the glenoid.

  • It is generally readily available and with minimal cost.

  • It also can be placed arthroscopically as well as used in both anterior and posterior cases of bone loss.

  • Although donor site morbidity has not been reported with this specific technique, graft harvest is similar to the Mumford technique, which has been reported to give excellent or good outcome in up to 85% of treated patients, with dissatisfaction correlated with clavicle over- or underresection. Researchers suggest excising 5 to 10 mm of a distal clavicle to be a safe method, which is similar to our technique ( ).

  • Recent work by the authors demonstrated that the DCA may on average reproduce up to 44% of the glenoid radius, which compares favorably to a 31% restoration when a traditional coracoid transfer is used ( ).

  • have reported supporting biomechanical data on this technique, describing that contact pressure differences between clavicular grafting and congruent arc coracoid transfer are favorable to the DCA.

  • Moreover, the distal clavicle graft is capped with articular cartilage that is within 1 mm of native glenoid cartilage thickness. It is a fresh, unprocessed tissue source that is immediately transplanted, so concerns about chondrocyte viability, immunorejection, and infection are minimized (Kwapisz Tokish, 2017).


  • Although promising anatomic results have been reported, there is a lack of clinical outcomes with this technique.

  • It does not augment or address anterior capsular structures that are often a part of complex instability cases.


  • It is the preference of the senior author to use the DCA in young patients who have glenoid bone loss as the primary reason for their instability, with defects from 15% to 30%, and relatively preserved soft tissue structures.


  • In cases of collagenopathies such as Ehlers-Danlos syndrome, previous thermal capsulorraphy, or multiply operated patients, we prefer an alternative technique that addresses these issues with soft tissue reinforcement.

Authors’ Preferred Surgical Technique

Preoperative Preparation

Patients with glenohumeral instability should undergo a standard history and physical examination, as well as preoperative advanced imaging such as CT or MRI. Glenoid bone loss is calculated in every patient, and this calculation aids in determining the operative approach to the patient according to the “on-track–off-track” concept. As relative indications for bony augmentation of either anterior or posterior instability, bone loss of greater than 15% of the glenoid diameter or the existence of significant retroversion in the presence of posterior instability is considered. Other factors such as age, athletic status, capsular laxity, and patient preferences are weighed when deciding among different treatment options.

Arthroscopic Portal Positioning

After the induction of general anesthesia, examination under anesthesia to confirm the preoperative diagnosis is performed. The patient is positioned in the lateral decubitus position on a beanbag with a padded axillary roll, with the use of a padded arm sleeve (STAR sleeve; Arthrex, Naples, FL), with balanced suspension.

A standard posterior portal is established approximately 1 cm medial and 2 cm distal to the posterolateral acromial border. The arthroscope is introduced, and additional portals are established using an outside-in technique under direct visualization with the use of a switching stick. The anterosuperior portal is established first, approximately 1 cm inferior to the clavicle and lateral to the coracoid. The midglenoid portal is created just superior to the superior border of the subscapularis. In cases of posterior augmentation, a 7 o’clock portal is created approximately 4 cm off the posterolateral corner of the acromion, bisecting the angle created by the posterior and lateral borders of the acromion, respectively, under direct visualization. To allow efficient switching of the camera and instruments throughout the case, 8.25-mm cannulas are liberally used.

Diagnostic Arthroscopy and Biologic Preparation

After the diagnostic arthroscopy is performed with particular attention to the pathology, the arthroscope is switched to the anterosuperior portal, and a 3-mm graduated probe is placed to confirm our preoperative measurements of glenoid bone loss. Biologic preparation includes a wide release of the glenoid labrum to ensure its mobility for accurate reduction, especially when the bone block reconstitutes the glenoid shape. This is performed with arthroscopic liberators and ablators. The glenoid is also biologically prepared with either an arthroscopic rasp or high-speed cylindrical bur, with the goal to create a healthy bed of bleeding cancellous bone, as well as to create a flat surface perpendicular to the glenoid surface to ensure a flush fit with graft placement.

Graft Harvest

A single 3-cm horizontal incision is made over the subcutaneous border of the acromioclavicular joint, along the midline of the clavicular axis. The skin and subcutaneous tissues are divided, and thick periosteal flap is raised to expose the joint and approximately 1 cm of the distal clavicle. A 1-cm-wide saw blade is used to remove the distal 1cm of clavicle, and soft tissue is cleaned from around the bone. The graft is placed on the back table, and the periosteal flap is closed with nonabsorbable #2 interrupted stitches. The remainder of the soft tissue is closed in two layers, and the wound is dressed at the completion of the case.

Graft Preparation

The distal clavicle is a versatile graft, with a variable amount of version and an articular surface that is generally 19 mm long 13 mm wide (Kwapisz Tokish, unpublished). The graft is evaluated based on its best fit and cut perpendicular to its articular surface to a width that matches the measurement of bone loss that was determined preoperatively and confirmed arthroscopically. In most cases, 7 to 8 mm of augmentation is normally sufficient to reconstruct up to 30% bone loss, and the graft is fashioned to anatomically fit and replace the loss. At this point, the method of fixation for the graft is chosen. If we decide on screw fixation, the graft is predrilled in a lag construct with pilot holes allowing compression of the autograft with the screw. Alternatively, we often use suture anchors to secure the graft; in these cases, three 1-mm holes are drilled in a triangular formation, with two drill holes, 1 mm in diameter, 3 to 4 mm off the articular surface, at the superior and inferior borders of the graft. The third hole is drilled medial with respect to the graft’s final position on the glenoid.

Delivery And Fixation Of Graft

Screw Fixation

If the graft is to be fixed with screw fixation, it can be passed either freely into the joint or along a K-wire guide predrilled in the glenoid. The advantage of a free pass is that the graft may fit down a standard midglenoid cannula and, after it has been inserted, can be flipped 90 degrees and advanced through the rotator interval inferiorly to match its resting position at the anterior-inferior glenoid, where it can be held in place with a liberator introduced from the posterior portal for anterior bone augmentation. Trying to pass the graft down a K-wire will require a wider exposure through the subscapularis to obtain a proper position. Likewise, the graft can be introduced through a posterior cannula and held in place with a liberator from the midglenoid portal. When in place, a K-wire is placed through the pilot holes of the clavicle graft and advanced into the native glenoid. This is usually not difficult for posterior grafts, but with anterior screw placement, the standard midglenoid portal may not be sufficient to achieve the appropriate angle. In such situations, an additional 5 o’clock portal is established through the subscapularis to ensure the correct trajectory. Extreme care is taken to protect the axillary nerve, and if any doubt exists, it is dissected arthroscopically, visualized, and protected. After the graft is secured to the glenoid in the appropriate position with the K-wire, a cannulated drill is advanced into the glenoid to allow lag fixation of the graft, with a cannulated, titanium 3.75-mm screw (Arthrex). If the graft is too large to easily be delivered, the cannula can be removed, the portal expanded, and the graft delivered directly. If the proper trajectory cannot be achieved with wire provisional fixation, then one can consider using a suture anchor as an alternative or conversion to an open approach.

Suture Anchor Fixation

If suture anchor fixation is selected, the previously drilled holes in the graft are noted by their measurements from the articular sur face and from each other. From these measurements, two 3.0-mm BioComposite SutureTaks (Arthrex) are placed at the superior and inferior borders of the bone defect at the corresponding distances from the articular surface and each other, respectively. All limbs are delivered out of the working portal. One limb from each suture anchor is passed through the medial “conjoined” drill hole. The other two sutures are passed, one each, through the superior and inferior articular drill holes. These latter sutures are tied in a square-knot fashion over the intervening bone bridge with three stacked half-hitches. The excess suture is not cut. Graft delivery is then accomplished through the cannula by a “double-pulley” technique whereby the free limbs are pulled, which brings the graft to the suture anchor eyelets because of the knotted ends of the opposite limbs of suture. Either the graft can be assisted with a switching stick through a cannula or, if the graft is too large, the portal can be enlarged slightly, and the graft can be introduced with the assistance of a curved hemostat. After the slack is pulled out of the anchor system, an arthroscopic knot with three additional half-hitches is tied, and the graft is secured to the native glenoid across two bone bridges in a “double-row” fashion ( Fig. 1H.1 ).

Fig. 1H.1

Distal clavicular osteochondral graft (DCA) after its placing on the glenoid ridge and fixing in a “double-row” fashion sutures. GS, Glenoid surface; HH, humeral head.

Incorporation of Native Labrum Into the Graft

The remaining tails are passed through the native labrum to bring it up to the neoarticular surface with the aid of retrograde suture lassos and tied down with secondary similar knots ( Fig. 1H.2 ). If screw fixation has been used, supplemental suture anchors can be placed either through grafts of larger size or at the superior and inferior borders of the graft if there is concern about there being enough accommodating graft. All arthroscopic instrumentation is removed, and the skin is closed and dressed sterilely.

Fig. 1H.2

Distal clavicular osteochondral graft (DCA) after its fixation to the deficient glenoid. Sutures are tied in a “double-row” fashion, and the remaining tail can be used to reattach native labrum. GS, Glenoid surface; HH, humeral head.

Postoperative Rehabilitation

The patient is placed in a neutral rotation sling for 6 weeks. Pendulums are allowed immediately, and passive motion is started at 3 weeks, with a goal to obtain full range of motion by 8 weeks. At 8 weeks’ follow-up, imaging is obtained, and if the graft looks incorporated, active motion is begun. Strengthening is added at 4 months postoperatively, and return to full activity is assessed at 6 months. Final radiographs are obtained at this point to ensure graft incorporation.


  • Glenoid bone loss can be addressed by a variety of different techniques.

  • Each has unique advantages and limitations.

  • This chapter has detailed the use of the distal clavicle osteochondral graft. This autograft provides a readily available and almost noncost method for anatomic reconstruction of glenoid bone loss.

  • The graft restores both the radius of the native glenoid and comparable amount of its native cartilage thickness.

  • It also compares favorably to the coracoid in terms of arc of restoration, providing a corticocancellous buttress for glenoid restoration.

  • Although this graft provides promising theoretical, anatomic, and biomechanical promises, longer term clinical studies are necessary to validate its use in the clinical setting.

Glenoid Bone Loss: Arthroscopic Bone Block

Ettore Taverna, MD
Vincenzo Guarrella


Shoulder dislocation can cause a variety of pathologic lesions. Patients with chronic anterior shoulder instability may present with recurrent dislocation, subluxation, or chronic shoulder pain. Anteroinferior glenoid bone deficiency has been reported in 22% of initial traumatic anterior shoulder dislocations and in up to 90% of recurrent anteroinferior shoulder instability cases. Many different open and arthroscopic techniques have been described to address this pathology. The benefits of using arthroscopic procedures for surgical stabilization of the shoulder include smaller incisions with less soft tissue dissection, better visualization of the joint, improved repair accessibility, and the best potential outcome for external rotation. We have arthroscopically treated a subgroup of patients affected by bone loss in recurrent anterior shoulder instability with a modified Eden-Hybinette technique since 2005. The procedure allows an accurate positioning of the bone graft (autograft or allograft) flush to the anteroinferior glenoid and a standard Bankart repair leaving the graft extraarticular. In our series, we found promising clinical and radiologic results.

Keywords: arthroscopic Eden-Hybinette; Arthroscopic guide; bone block; EndoButtons; glenoid bone loss; shoulder instability.


  • Shoulder dislocation can cause a variety of pathologic lesions. Patients with chronic anterior shoulder instability may present with recurrent dislocation, subluxation, or chronic shoulder pain ( ).

  • Anteroinferior glenoid bone deficiency has been reported in 22% of initial traumatic anterior shoulder dislocations and in up to 90% of recurrent anteroinferior shoulder instability cases ( ). Many different open and arthroscopic techniques have been described to address this pathology ( ).

  • The benefits of using arthroscopic procedures for surgical stabilization of the shoulder include smaller incisions with less soft tissue dissection, better visualization of the joint, improved repair accessibility, and the best potential outcome for external rotation ( ).

  • In 2005, we developed an arthroscopic technique that allows an accurate positioning of the bone graft (autograft or allograft) flush to the anteroinferior glenoid and a standard Bankart repair leaving the graft extraarticular. In November 2013, we made our last modification consisting in fixation using two arthroscopic EndoButtons (instead of two screws) placed through a guide; optimal compression is then obtained with a tensioning device ( ).The guide permits parallel drilling of the two tunnels perpendicular to the glenoid neck, reducing the risk of graft resorption and nonunion ( ). Other advantages include a quicker rehabilitation with early return to sport activities and a better cosmetic result. In our series, early clinical and radiologic results are promising.


Our indication to perform an arthroscopic bone-block procedure in the setting of recurrent anterior instability is young age (younger than 45 years old), maximum of five dislocations, first dislocation within 3 years, good radiologic quality of soft tissues, and glenoid bone deficit between 15% and 25% on preoperative computed tomography (CT) scans (according to ). Other indications include arthroscopic evidence of significant bone loss ( ) or glenoid bone loss between 10% and 20% associated with a Bankart lesion or an instability severity index score (ISIS) ( ) between 3 and 6. Other indications include failure of previous isolated soft tissue repair or Bristow-Latarjet procedure. In our practice, indications for arthroscopic soft tissue repair are ISIS less than 3 and no glenoid bone loss, isolated bone loss of 10% or less, and a maximum of three dislocations. We prefer to perform Latarjet procedures in patients with ISIS greater than 6, more than five dislocations, chronic instability for more than 3 years, and glenoid bone loss greater than 25%.

Surgical Technique

Patient Positioning and Joint Preparation

Under general anesthesia and with the administration of perioperative antibiotics, the patient is placed in the beach-chair position. The scapula can be bolstered to rotate the glenoid externally.

A standard posterior portal is created for the insertion of the arthroscope. Viewing from the posterior portal, the surgeon creates an anterosuperior portal and midglenoid portal, and two 5.5-mm cannulas are introduced in the rotator interval. Initially, the labrum is accurately detached from the glenoid rim, and all soft tissues are removed from the anterior glenoid neck using a soft tissue shaver. We then introduce the arthroscope through the anterosuperior portal, and the anterior glenoid rim is further decorticated and flattened with a motorized bur to create a flat and bleeding bony surface to accommodate the graft ( Fig. 1I.1 ). At this point, a spinal needle is inserted from posterior to anterior along, and perfectly parallel to, the face of the glenoid and centered on the anterior glenoid bone defect. Then a more posteromedial portal is made to provide access for the glenoid guide ( ).

Fig. 1I.1

Arthroscopic view from posterior portal. The labrum and capsule are elevated, and the anterior glenoid rim is decorticated to create a flat surface to accommodate the graft.

From Taverna E, D’Ambrosi R, Perfetti C, et al. Arthroscopic bone graft procedure for anterior inferior glenohumeral instability. Arthrosc Tech. 2014;3(6):e653-660.

  • Video 1I.1

    Glenoid bone loss: arthroscopic bone block.

Glenoid Guide and Drill Pin Placement

The hook end of the glenoid guide is inserted through the posteromedial portal. The hook is passed along the glenoid parallel to the glenoid face to avoid damaging the articular surface, and then it is passed over the anterior edge ( Fig.1I.2, A ). When sufficiently advanced, the guide is rotated to capture the anterior edge of the glenoid under the hook. The hook should be placed at the center of the anterior glenoid defect, usually between the 3 and 4 o’ clock positions ( Fig. 1I.2, B ). It is mandatory that the glenoid guide is aligned with the posterior and anterior glenoid rim. After the guide is positioned, a bullet is placed in the inferior hole of the guide ( Fig. 1I.3 ). A small skin incision is made, and the bullet is advanced until it firmly contacts the posterior aspect of the glenoid neck ( Fig. 1I.4 ). The ratchet teeth of the bullet should be aligned with the screws adjacent to the guide handle. The process is repeated for the superior bullet ( Fig. 1I.2, C ). A 2.8-mm sleeved drill is placed in each bullet and advanced under power until exiting from the anterior aspect of the glenoid. The drills are placed 5 mm on the center below the cortical edge of the glenoid face, parallel to one another and 10 mm apart ( Fig. 1I.2, D ). The inner drill is removed, leaving the cannulated outer sleeve. Arthroscopic fluid exiting from the outer sleeve posteriorly confirms intraarticular positioning ( Fig. 1I.2, E ).

Fig. 1I.2

Glenoid guide and drill pin placement. A, Insertion of hook from posterior portal. B, Rotation of hook to capture glenoid rim. C, Insertion and advancement of bullet to posterior border of glenoid. D, Positioning and advancing of sleeves. E, Removal of inner drill. F, Removal of bullet. G, Introduction of 10-mm cannula through rotator interval and retrieval of wires.

From Taverna E, D’Ambrosi R, Perfetti C, et al. Arthroscopic bone graft procedure for anterior inferior glenohumeral instability. Arthrosc Tech. 2014;3(6):e653-660.

Fig. 1I.3

Glenoid guide with special bullets used to create holes from posterior to anterior at level of glenoid rim and two 2.8-mm sleeves.

From Taverna E, D’Ambrosi R, Perfetti C, et al. Arthroscopic bone graft procedure for anterior inferior glenohumeral instability. Arthrosc Tech. 2014;3(6):e653-660.

Fig. 1I.4

Outside view of proper positioning of glenoid guide and insertion of the bullet (right shoulder).

From Taverna E, D’Ambrosi R, Perfetti C, et al. Arthroscopic bone graft procedure for anterior inferior glenohumeral instability. Arthrosc Tech. 2014;3(6):e653-660.

After drilling is completed, the bullets can be removed by rotating each bullet to disengage the ratcheting teeth and extracting them posteriorly ( Fig. 1I.2, F ). The guide can be removed at this stage. Care should be taken to ensure that the sleeves remain firmly positioned in the glenoid neck. Flexible looped guidewires are then introduced into the joint by passing one wire through each sleeve posterior to anterior ( Fig. 1I.5, A ). Each guidewire is retrieved using a loop grasper, which is passed through the 10-mm cannula introduced through the rotator interval ( Fig. 1I.5, B ). The wires are separated and stored. At this point, with the drill sleeves in place, drilling of anteroinferior glenoid rim should be performed for the placement of suture anchors for soft tissue Bankart repair (anchors will be inserted at the end of the procedure). The drill sleeves should be removed after this step is completed ( Fig. 1I.2, G ).

Fig. 1I.5

A, Arthroscopic view from the posterior portal of the flexible loop guidewires in the joint (Looped Guidewire). B, Arthroscopic view from posterior portal of loop grasper used to retrieve flexible loop (Loop Grasper).

From Taverna E, D’Ambrosi R, Perfetti C, et al. Arthroscopic bone graft procedure for anterior inferior glenohumeral instability. Arthrosc Tech. 2014;3(6):e653-60.

Bone-Block Preparation

The tricortical bone graft is harvested from the ipsilateral anterior iliac crest. After a 4-cm skin incision has been made and splitting of the subcutaneous tissue has been performed, the inserting muscles are dissected from the iliac crest, and 2.5 to 3 cm of bone is marked with a fine-tip cautery while keeping a minimum safety distance of 4 cm posterior to the anterior superior iliac spine. The tricortical iliac crest bone block, measuring 20 mm × 8 mm × 8 mm, is fashioned using the graft master preparation board. Two 2.8-mm drill holes are made 10 mm apart and 5 mm from each edge. The drill enters through the cortex and exits the cancellous side of the bone block. The holes created correspond to the distance of the cannulated drill sleeves previously placed in the glenoid neck. With a marking pen, we color the superior aspect of the bone block.

Graft Passage and Loading of Implants (Anterior and Posterior Round Endobuttons)

Before loading the implant onto the guidewires, care is taken to ensure that the looped guidewires are not tangled within the joint. Each looped guidewire is fed through the prepared bone block and exits on the cortical side ( Fig. 1I.6, A ). The bone block is oriented so that the cancellous surface is facing the anterior neck of the glenoid. The anterior implant is fed with the preassembled suture through the end of the looped guidewire with a classic slip knot. This can be achieved by passing the lead suture through the looped guidewire and feeding the implant through the lead suture ( Fig. 1I.6, B ). The bone block is slid toward the end of the guidewires to lodge the implants. Anterior round EndoButtons (Smith & Nephew, London, England) ( Fig. 1I.7 ) are advanced until they lie flat on the bone block. Sutures should be taut to allow smooth movement down the cannula ( Fig. 1I.6, C ). The bone block is tipped to be inserted into the 10-mm cannula, and care is taken to ensure that the superior end of the bone block enters the cannula first ( Fig. 1I.6, D ). The bone block is advanced by pulling the guidewires out posteriorly. Slight tension should be maintained on the sutures throughout this step ( Fig. 1I.6, E ). The sutures should advance the implant until the bone block sits flush on the anterior neck of the glenoid, with each implant’s lead suture exiting the skin posteriorly ( Fig. 1I.6, F ). The posterior implants are placed on the transporter by advancing the instrument through the eyelet of a posterior round EndoButton ( Fig. 1I.6, G ). The suture is passed through the transporter. The transporter is retracted to allow the suture to pass through the eyelet of the posterior round EndoButton. The same steps must be performed for the second eyelet with the other side of the suture ( Fig. 1I.6, H ). The posterior round EndoButtons are advanced until they sit flush against the posterior face of the glenoid. The knot pusher is used to secure the posterior round EndoButtons. The knot pusher will provide tactile feedback when the posterior round EndoButtons are properly seated ( Fig. 1I.6, I ).

Sep 14, 2018 | Posted by in SPORT MEDICINE | Comments Off on Anterior Shoulder Instability
Premium Wordpress Themes by UFO Themes