Open Coracoid Transfer: Indications, Technique, and Results

Fig. 15.1

Effects of the Latarjet procedure: (1) increased glenoid surface area because the coracoid graft; (2 e 3) slinglike construct of subscapularis and common tendon that create a dynamic reinforcement to the deficient anteroinferior capsulolabral complex

Since the Latarjet technique was developed well before recent arthroscopic capsuloplasty procedures, it follows that the indications for “coracoid transposition,” although clearly considered a surgical procedure to treat inferior trauma-induced instability, may still differ slightly depending on the particular school of surgery, international scientific society, or geographical area.

It should be noted, however, that although the consensus is for “coracoid transposition” surgery (hereinafter called the “Latarjet technique”) to be performed on patients with substantive glenoid bone loss, some surgical schools also include instability with minimum bone loss and poor quality soft tissue in their indications for surgery.

Many studies have evaluated the contributing factors for recurrent instability following nonoperative and operative treatment (arthroscopic capsulolabral repair/Bankart repair) of the anterior glenohumeral instability.

Risk factors for recurrent instability include young age [12, 13], anterior glenoid bone loss [14] or posterior humeral [15] bone loss (Hill-Sachs), underlying ligamentous laxity or multidirectional instability [16], prior ipsilateral anterior shoulder dislocation [16, 17], neurologic deficit or voluntary instability, prior or concurrent ipsilateral rotator cuff tear [18], and participation in contact or collision sports [13, 19–22].

In cases of anterior instability in which substantial glenoid bone deficiency is observed, an isolated Bankart repair is unlikely to result in a stable shoulder. Patients with glenoid bone loss measuring ≤15% can typically be treated with rehabilitation or soft tissue procedures alone [23]. For patients with an intermediate amount of bone loss (15–30% of the glenoid surface), it is important to consider the patient’s demands on the shoulder when formulating a treatment plan. Low-demand patients may be treated successfully with either conservative methods or soft tissue procedures. However, high-demand patients will likely benefit from addressing the glenoid lesion to prevent recurrent instability [23]. For nearly all patients with bone loss measuring a ≥25% to 30% of the glenoid surface, a surgical procedure addressing the glenoid bone loss is necessary to prevent further instability [24].

Although these guidelines have been reflected in the literature, efforts should be made to reduce and fix displaced fracture of the glenoid rim in young, active patients to improve postoperative stability [25].

Burkhart and De Beer [14] recognized that one of the risk factors for failure of arthroscopic stabilization was based on the anatomic relation of the bone loss affecting the humeral head and the glenoid in critical positions. In fact, they introduced the concept of “significant bone loss.” They defined a significant glenoid bone defect as in which the arthroscopic appearance of the glenoid, when viewed from a superior-to-inferior perspective, was an inverted pear. On the humeral side, they defined a significant bone defect to be an engaging Hill-Sachs lesion, oriented in such a way that it engaged the anterior glenoid in a position of athletic function (90° of abduction combined with external rotation of approximately 90°). They found that the instabilities associated with “engaging-type” Hill-Sachs lesions were at high risk of recurrence if treated with the classic arthroscopic capsuloligamentous repair, confirming that the restoration of the soft tissues alone would not be sufficient to contain the humeral head under stress.

However, this diagnostic technique could potentially cause an overdiagnosis of engaging Hill-Sachs lesions because ligament insufficiency (Bankart lesion) might permit the humeral head to excessively translate anteriorly, thus facilitating engagement of the humeral defect with the glenoid rim [26].

The importance of bone loss on both the glenoid and humeral side has been increasingly studied; for the purpose of evaluating the size of the Hill-Sachs lesion together with the size of the glenoid, the “glenoid track” concept was introduced by Eiji Itoi [27].

The glenoid track is a contact zone of the glenoid on the humeral head with the arm at the end range of motion, e.g., in various degrees of elevation with the arm in maximum external rotation and maximum horizontal extension. This end range of motion is critical for anterior dislocation because the anterior soft tissue structures become tight and prevent the anterior translation of the humeral head in this position. It is this position that patients with recurrent anterior dislocation of the shoulder feel anterior apprehension. If the Hill-Sachs lesion is always covered by the glenoid at this end range of motion or, in other words, if the Hill-Sachs lesion stays within the glenoid track, the lesion does no harm, because it is always covered by the glenoid even at the end range of motion. On the other hand, if the lesion comes out of the glenoid coverage, it engages with the anterior rim of the glenoid and causes a dislocation. Clarifying the exact location of this contact zone or the glenoid track enables to evaluate any Hill-Sachs lesion for its risk of engagement [28].

A more recent evolution considers how both the glenoid and the humeral bone loss interact to determine whether their combination results in an “on-track” or “off-track” lesion, which may be more predictive of recurrent instability than looking at either side individually [29].

A method has been developed (both radiographic and arthroscopic) that uses the concept of the glenoid track to determine whether a Hill-Sachs lesion will engage the anterior glenoid rim, whether or not there is concomitant anterior glenoid bone loss. If the Hill-Sachs lesion engages, it is called an “off-track” Hill-Sachs lesion; if it does not engage, it is an “on-track” lesion. On the basis of this quantitative method, a treatment paradigm with specific surgical criteria for all patients with anterior instability, both with and without bipolar bone loss, can be applied.

Regardless of the degree of bipolar bone loss, anterior instability can be categorized as follows:

Group 1, glenoid defect of less than 25% plus on-track Hill-Sachs lesion – treatment: arthroscopic Bankart repair
Group 2, glenoid defect of less than 25% plus off-track Hill-Sachs lesion – treatment: arthroscopic Bankart repair plus remplissage
Group 3, glenoid defect of 25% or more plus on-track Hill-Sachs lesion – treatment: Latarjet procedure
Group 4, glenoid defect of 25% or more plus off-track Hill-Sachs lesion – treatment: Latarjet procedure plus humeral-sided procedure (humeral bone graft or remplissage) if the Hill-Sachs lesion is engageable by surgeon on operating room table after Latarjet procedure or only Latarjet procedure if Hill-Sachs lesion is not engageable by surgeon after Latarjet procedure

Numerous alternative sources of bone graft have been used to address the glenoid bone deficiency. Historically these have been osseous autografts from the iliac crest as described by Eden in 1918 and Hybbinette in 1932. They described the use of an L-shaped iliac crest bone block placed anterior to the glenoid under the periosteum, leaving the “L” arm of the graft extending lateral to the anterior glenoid. This was modified by Alvik who secured the graft into a preformed wedge-shaped groove on the anterior glenoid and later by De Palma who used screw fixation to secure the graft [30, 31]. Despite reports of positive outcomes, recurrent instability and a high incidence of moderate or severe arthritis developed with longer follow-up [32, 33].

More recently Warner reported on the anatomical fixation of autologous iliac crest using screws in 11 patients that had glenoid bone loss. Despite 9 of 11 cases having prior unsuccessful operative procedures at a mean follow-up of 33 months, all had improved ASES and Rowe scores with all patients returning to their pre-injury level of sport [34].

The Lyon group has recommended the use of the modified Eden-Hybbinette procedure for recurrent anterior dislocation after failed Latarjet with 79% good or excellent outcomes [35]. Arthroscopic-assisted procedures have recently been described using autologous iliac crest to restore the anterior bone loss using either screw fixation or a J-shaped press-fit technique [36, 37].

An alternative autologous option is the use of an ipsilateral distal clavicle resection which provides favorable osteochondral characteristics as a replacement for glenoid bone [38]. No clinical studies have yet been reported.

Current coracoid transfer procedures are nonanatomic, do have harvest morbidity, and are associated with an increasing number of complications. In a review of 45 level IV studies reporting outcomes following Bristow or Latarjet shoulder stabilization procedures, a total complication rate of 30% was reported [7].

Allograft reconstruction of glenoid bone loss represents a potential alternative to autologous coracoid use. There are numerous sources of allograft being used including glenoid, iliac crest, distal tibial plafond, femoral head, and humeral head. Some are osteochondral grafts and therefore theoretically provide both bone and hyaline cartilage surfaces. Data from studies using allograft for recurrent shoulder instability show excellent clinical outcomes, a low rate of recurrence, good graft union, and low rates of graft resorption [39]. All included studies were of level IV evidence, and the likelihood of methodological bias is increased. Unfortunately to this date, no comparative studies between the use of coracoid transfer and allograft for shoulder instability exist.

All of these alternative bone graft procedures restore stability by increasing the glenoid surface area alone and obviously do not have any sling effect from the conjoined tendon and inferior subscapularis.

15.2 Technique

15.2.1 Surgical Approach

The patient is instructed to shave the shoulder girdle and take a shower using antibacterial soap the night before surgery. Immediately before surgery, an interscalene block is placed for postoperative pain control. A general anesthetic is administered, and the patient is placed in the modified beach chair position with a 1-cm thick folded sheet placed under the scapula on the affected side, making the coracoid process readily palpable.

The deltopectoral approach is routinely used for the open treatment of anterior glenohumeral instability, regardless of the procedure that will be performed.

A 5-cm skin incision is made starting at the tip of the coracoid process and extending inferiorly.

The deltopectoral interval is located superiorly and medially by identifying the small triangular area devoid of muscle. The cephalic vein is identified in the deltopectoral interval and ligated with braided absorbable suture to prevent postoperative hematoma; the intermuscular plane is developed and retracted with right angle retractors, taking the cephalic vein laterally. Retraction of the vein medially carries the risk of injuring the large veins that drain the deltoid muscle.

A self-retaining retractor is then placed between the pectoralis major and the deltoid, completing the operative exposure (Fig. 15.2).

Fig. 15.2

The coracoid graft

The arm is abducted and externally rotated. Mayo scissors are used to clear the superior aspect of the coracoid process, and a Hohmann retractor is placed over the top of the coracoid process. The conjoined tendon is identified, and the coracoacromial ligament and the coracohumeral ligament are completely transacted at its lateral coracoid insertion. The arm is placed in an internally rotated position, and the pectoralis minor tendon is identified and released from its coracoid insertion taking care not to disturb the blood supply to the coracoid process, which enters at the medial aspect of the coracoid insertion of the conjoined tendon (Fig. 15.3). After release of the pectoralis minor tendon, a periosteal elevator is used to expose the “knee” of the coracoid process by sliding it along its medial aspect. Cutting of the coracoid at the level of the “knee” is initiated with a microsagittal saw equipped with a 90° angled blade and completed with an osteotome. Grasping forceps are used to hold the coracoid process gripping the medial and lateral aspects; the arm is returned to the abducted and externally rotated position.

Fig. 15.3

Pattern of the coracoid blood supply

A gauze sponge is placed over the skin at the distal aspect of the incision, and the coracoid process is placed on the sponge by flipping it (the deep surface should be superficial and the superior aspect should be distal); any remaining soft tissue is removed from the deep surface of the coracoid process [40].

The microsagittal saw and high-speed burr are then used to decorticate the deep surface of the coracoid bone graft; sterile saline solution is useful to reduce the heat caused by the saw during cutting. Exposing the cancellous bone will indicate that decortication has been performed correctly, and bleeding from cancellous bone will improve integration of the bone graft on the glenoid neck.

A drill guide allows the surgeon to create two dorsal-ventral holes in the coracoid graft using the 2.7-mm drill perpendicular to its long axis (if malleolar screws are intended to be used, the drill is 3.2 mm). The guide has a distance between centers equal to that of the wedged profile plate if intended to be used to improve bone-to-bone compression. The parallel positioning of the two wires guarantees that the two screws will be positioned perfectly parallel [41].

The lateral border of the conjoined tendon can be further released to additionally mobilize the coracoid process if necessary, and the coracoid process is placed beneath the arm of the self-retaining retractor holding the pectoralis major.

The subscapularis tendon and muscle is exposed with the arm by the side and externally rotated. The superior and inferior margins of the subscapularis should be identified.

The subscapularis muscle is divided in line with its fibers using Mayo scissors. Normally, the level of division is the junction of the middle and inferior thirds of the muscle; however, in the case of hyperlax patient, the junction of the superior and inferior half is selected to maximize the effect of the conjoined tendon sling and of the inferior sub-scap. The scissors are opened vertically, exposing the underlying capsule. To facilitate opening of the scissors and capsular exposure, it may be necessary to decrease the amount of external rotation of the arm. Taking down the subscapularis insertion (vertical tenotomy), whether partially or fully, requires protection of passive external rotation for at least 6 weeks in addition to graduated internal rotation strengthening program [42]. A subscapularis split may result in significantly less morbidity while also allowing the subscapularis to function as a “sling” component of the Latarjet reconstruction [42]. Once the capsula is well visualized, heavy forceps are used to develop the plane between the anterior surface of the scapula and the subscapularis muscle belly, allowing placement for a gauze sponge in the subscapularis fossa and elevating the subscapularis muscle off the capsule. A Hohmann-type retractor is placed on the anterior surface of the neck of the scapula as far medial as possible. The inferior portion of the subscapularis is retracted inferiorly, and using a scalpel, the lateral portion of the subscapularis is divided in line with its fibers to its insertion on the lesser tuberosity. The well-visualized underlying capsule allows vertical (close to the glenoid) or horizontal capsulotomy that will facilitate placement of a humeral head retractor into the glenohumeral joint. At this point, two retractors are used: a standard Fukuda retractor laterally for the humeral head and a forked glenoid-type retractor placed on the anterior scapular neck as medially as possible to improve visualization of the site of graft insertion. The superior portion of the subscapularis muscle is then retracted superiorly and held by a Steinmann pin driven into the surgical neck of the scapula (or by dedicated retractor). A small Hohmann retractor placed under the scapular neck could be useful.

The anteroinferior glenoid rim surface is cleared of soft tissue. With a high-speed burr, the anteroinferior glenoid neck is prepared for placement of the coracoid bone graft. The good exposure of the glenoid, obtained with the use of the specific retractors, makes it possible to obtain the best view of the surgical field. The anterior-inferior region of the glenoid neck is prepared to obtain bleeding from the bone and to attain correct leveling of the surface to aid the bone-to-bone contact between glenoid and coracoid graft.

At this point, the positioning of the coracoid is decided. Our criterion is biological and biomechanical [43]. Since the inferior part of the coracoid nearest the conjoined tendon is the most vascularized section, we tend to insert it where the greatest “bone loss” has occurred – usually at 3 and 5 o’clock – in order to allow optimal mechanical stimulation of the graft (mechanotransduction) (Fig. 15.4a, b).

Fig. 15.4

(a) CT en face view with subtraction of the humeral head is evident osteolysis of the coracoid graft when glenoid bone loss is mild. (b) CT en face view with subtraction of the humeral head is evident good integration of the most inferior part of the coracoid graft when glenoid bone loss is large

The first guide wire (a 0.9-mm K-wire with threaded apical region) is inserted for subsequent drilling of the holes through the glenoid. The first wire to be inserted is the lower one in order to obtain correct positioning of the lower screw and therefore of the entire implant (wedge plate).

It is important to position this guide wire while precisely observing the lower glenoid margin to ensure that the distal screw will not be implanted too far down.

The coracoid graft offset is crucial; generally speaking, the coracoid has a lateromedial surface of approximately 2 cm; therefore, the first wire will be implanted approximately 1 cm medially from the glenoid margin. The guide wire must be positioned parallel to the joint surface of the glenoid to avoid positioning the screws inside the joint in the following phase, which would cause serious damage to cartilage surfaces.

The second 0.9-mm K-wire with its guide is positioned, with indication of the offset if used.

To obtain better compression and load distribution between the coracoid graft and glenoid bone surface, a miniplate can be used (wedged profile plate; Arthrex Inc., Naples, FL, USA), the characteristics of which each correspond to a specific biomechanical function appropriate for the Latarjet procedure. The figure-of-eight configuration allows for better torsional orientation of the plate on the dorsal coracoid bone graft surface. This allows the plate to distribute the load evenly throughout the bone, avoiding stress risers that occur when screws only are used. When compressed during the screwing phase, the miniplate, positioned with the wedge profile oriented medially, will tilt the coracoid bone graft in that direction, aligning the bone contact surface between the coracoid graft and the steep glenoid neck. Four spikes on the plate are designed to hold the graft as a whole and to reduce traction forces from the conjoined tendon, at the same time improving plate and graft stability during surgical fixation: plate-coracoid-glenoid neck. The more the bone loss, the less steep is the glenoid neck. In this situation, the improved compression force effect from the plate is more important than the wedge effect; the wedge is useful in minor bone loss because of the steeper glenoid neck.

Only gold members can continue reading. Log In or Register to continue