The scapulothoracic joint is not a true synovial joint. Rather, the scapulothoracic articulation is formed by the convex surface of the posterior thoracic cage and the concave surface of the anterior scapula. The scapula is a flat bone, with the gliding surfaces formed by the subscapularis and the serratus anterior. It is attached to the axial skeleton through the acromioclavicular joint and the sternoclavicular joint. The scapulothoracic articulation allows increased shoulder elevation. For every 2º of glenohumeral elevation, there is 1º of scapulothoracic elevation.
There are bony projections that serve as attachments for muscles and other soft-tissue structures.  These projections include the following:
Lateral scapular spine
The acromioclavicular ligament connects the distal end of the clavicle to the acromion and provides horizontal stability. The coracoclavicular ligament is made up of two bands: the conoid and the trapezoid, both of which provide vertical stability. The coracoacromial ligament connects the coracoid process to the acromion.
There are two major bursae: the scapulothoracic bursa, between the serratus and the thorax, and the subscapularis bursa, between the subscapularis and the serratus.
There are 17 muscles that attach to or originate on the scapula (see Table 1 below), and they perform two major roles: (1) to maintain a stable base of support for the humerus and (2) to allow for dynamic positioning of the glenoid fossa during glenohumeral elevation. The scapula can rotate upward and downward, can protract and retract, and can elevate or depress.
Table 1. Muscles of Scapula (Open Table in a new window)
Adducts, elevates, rotates,
and depresses the scapula
|Serratus anterior||Long thoracic||Protracts and rotates the scapula upward; maintains the medial angle of the scapula against the chest wall|
|Deltoid||Axillary||Abducts, adducts, flexes, and extends the arm|
|Latissimus dorsi||Thoracodorsal||Adducts, extends, and internally rotates the humerus|
|Levator scapulae||Third and fourth cervical||Elevates the scapula|
|Rhomboid major||Dorsal scapula||Adducts the scapula|
|Rhomboid minor||Dorsal scapula||Adducts the scapula|
|Pectoralis major||Medial and lateral pectoral||Adducts and internally rotates the shoulder joint and assists in forward elevation|
|Pectoralis minor||Medial pectoral||Protracts and rotates the scapula inferiorly|
|Teres major||Lower subscapular||Adducts and internally rotates the arm|
|Triceps brachii||Radial||Extends the forearm|
|Biceps brachii||Musculocutaneous||Flexes and supinates the forearm|
|Coracobrachialis||Musculocutaneous||Flexes and adducts the arm|
|Infraspinatus||Suprascapular||Externally rotates the humerus|
|Subscapularis||Upper and lower subscapular||Internally rotates the humerus|
|Teres minor||Axillary||Externally rotates the arm|
|Supraspinatus||Suprascapular||Abducts the humerus|
The scapula upwardly rotates in the frontal plane, posteriorly tilts in the parasagittal plane, and externally rotates in the transverse plane during functional elevation. Scapular control is essential to scapulohumeral coordination. Posterior tilting is responsible for humeral clearance during the acromiohumeral portion of shoulder elevation.
Fung et al discovered that scapular upward rotation and retraction are greatest during abduction elevation, as compared with flexion elevation.  They also discovered that posterior tilting was greatest during flexion elevation. Any disturbance in this rhythm can decrease scapulothoracic movement and can be associated with fatigue, impingement, instability, and limits in elevation.
Disorders of the scapulothoracic joint are not very common. The major disorders include the following:
Snapping scapula syndrome (scapulothoracic crepitus)
Facioscapulohumeral muscular dystrophy (FSHMD)
The term dyskinesis has been used to describe abnormal position or motion of the scapulothoracic joint; it can be caused by pain, muscle weakness, muscle inflexibility, or muscle imbalances. Any process that affects the scapulothoracic joint can affect the overall function of the shoulder joint and may present as posterior shoulder pain, periscapular pain, rotator cuff bursitis, or tendinitis secondary to impingement.
Snapping Scapula Syndrome
Snapping scapula syndrome (scapulothoracic crepitus) is a disorder that ranges from benign to disabling. [4, 5, 6, 7, 8, 9, 10] Scapulothoracic motion produces a snapping, grinding or popping sensation.
Many causes have been suggested for this syndrome, ranging from repetitive forceful shoulder movements producing microtrauma, resulting in a bone spur at the muscular attachment on the scapula, to crepitus that is the end result of bursitis. Milch proposed another theory: that crepitus may originate from bursitis but that the resulting pathology is caused by a soft-tissue lesion, such as an osteochondroma. 
Among the reported causes of snapping scapula are bony alterations and soft-tissue reactions.  Occasionally, there is no identifiable cause. A study by Patzkowski et al  proposed that snapping scapula syndrome may be more prevalent in the military as a consequence of intense upper-body exercise, prolonged heavy load-bearing through the shoulder girdle, and forces applied to the scapulothoracic articulation by protective gear.
Structural abnormalities that can lead to snapping scapula include the following:
Bony prominences, such as the Luschka tubercle (the tubercle of Luschka is an exostosis with bony enlargement of superomedial scapula, or scapula hook)
Other bony alterations include the following:
Abnormal curvature of the superior angle of the scapula
Sprengel deformity, which is a complex anomaly associated with malposition and dysplasia of the scapula, with muscle hypoplasia or atrophy causing disfigurement and limitation of shoulder movement
Curling of the vertebral border
Irregularities of the subscapular ribs
Exostosis of the subscapular ribs
Snapping scapula may be caused by muscle atrophy and nerve injury, which decreases the amount of soft tissue between the scapula and the rib cage, or fibrosis from an injury. Another possibility is bursitis or lesions from tuberculosis or syphilis, but this is unlikely. Other soft-tissue causes include the following:
Exostosis bursata, which is a bursa formation associated with an osteochondroma
Interstitial myofibrosis of surrounding muscles (or imbalance)
The diagnosis of snapping scapula is based on scapular noise with motion, and the patient may complain of pain and fatigue with activities. On physical examination, some patients may have tenderness to palpation along the scapula border; others may have no tenderness. Patients may also have scapular winging (see Scapular Winging below).
In addition to winging, the clinician must assess the strength and flexibility of muscles surrounding the shoulder girdle. Cervical radiculopathy at the C5-8 levels has to be ruled out as a possible cause of shoulder pain. A loss of scapulothoracic rhythm can lead to increased friction, which will produce crepitus. [6, 7]
Ancillary tests that help with the diagnosis of snapping scapula include anteroposterior (AP) and Y-view radiography and three-dimensional (3D) computed tomography (CT) to evaluate the bony incongruity between the anterior scapula and the chest wall. If there is no bony evidence for the crepitus, the next step would be magnetic resonance imaging (MRI) to evaluate for soft-tissue pathology (eg, an inflamed bursa). Electromyography (EMG) and nerve conduction studies are important for ruling out neurologic causes of dysfunction.
Treatment must focus on correcting muscle imbalances, strengthening weak muscles, stretching tight muscles, and correcting poor postures.
Physical therapy modalities can be beneficial for pain relief. It is extremely important not only to strengthen the stabilizers but also to work on endurance training. If these conservative measures are not successful, then nonsteroidal anti-inflammatory drugs (NSAIDs) and fluoroscopically guided injection of a steroid–local anesthetic mixture may be tried. Injection techniques of the scapulothoracic bursa are not well defined in the literature but have been shown to be of diagnostic and therapeutic value.
A study by Hodler et al demonstrated that fluoroscopically guided scapulothoracic bursa injections not only confirm the diagnosis but also are beneficial for at least temporary pain relief lasting between 6 hours and 15 months.  Chang et al reported a series of 22 cases in which scapulothoracic bursa injections of steroids plus hyaluronate were administered.  A series of three weekly injections were completed, and at 3-month follow-up, significant improvements were noted.
Once pain and inflammation are controlled, functional activities should be addressed so that the patient can return to specific job duties or sport-specific maneuvers. If conservative treatment fails after about 4-6 months, referral for surgical options is appropriate. Surgical options depend on the specific cause and include the following [4, 8] :
Excision of the superomedial angle of the scapula
Combined bursectomy and superomedial angle resection
When prolonged conservative treatment has failed and pain and disability impair daily activities or ability to perform in the workplace, surgery should be considered. Richards et al reported three cases of painful scapulothoracic crepitus that were successfully treated by resecting the superomedial angle of the scapula. [8, 15] No postoperative complications occurred, and patients all returned to their occupations. During follow-up evaluation, patients remained free of symptoms and had no complaints of weakness, instability, or scapular winging.
Lehtinen et al  examined the following five methods of surgical decompression of the scapulothoracic articulation for scapulothoracic bursitis:
Arthroscopic scapulothoracic bursectomy
Open scapulothoracic bursectomy
Open resection of the scapulothoracic bursa with excision of the superomedial portion of the scapula
Arthroscopic resection of the scapulothoracic bursa with excision of the superomedial portion of the scapula
Combined procedure consisting of arthroscopic scapulothoracic bursectomy and open resection of the superomedial scapula through a small incision
Each of the patients complained of periscapular pain with associated tenderness under the superomedial portion of the scapula. They also complained of pain during movement of the shoulder, which imposed various limitations of daily activities.
The vast majority of patients in the Lehtinen study had chronic pain and an audible crepitation or snapping.  Clinical outcomes at postoperative follow-up were encouraging. Eighty-one percent of patients reported satisfactory pain relief and answered yes when questioned if they would undergo the procedure again. There appeared to be no complications after the procedures, and there was no evidence of muscular detachment. There appeared to be no differences in outcome among the five methods used during decompression.
In 2014, Warth et al published a critical review of current evidence.  The 81 relevant, unique articles reviewed included 26 case series of fewer than 10 patients, 16 technique papers, 11 imaging studies, nine anatomic studies, and nine level IV outcomes studies. Because the level of evidence was not high enough to permit a meta-analysis or systemic review, a critical review was performed.
The authors concluded that when nonoperative therapy fails, bursectomy with or without partial scapulectomy is currently the most effective primary treatment.  Even with surgical treatment, however, many patients still experience shoulder disability. Furthermore, because the precise origin is typically unknown, specific treatments that are effective for some patients may be ineffective for others. The authors suggested future studies focusing on identifying modifiable factors associated with poor outcomes after conservative and surgical management.
There are several causes of scapular winging. Injury to the long thoracic nerve can produce medial winging, and weakness in the trapezius secondary to spinal accessory nerve injury can produce lateral winging. Isolated trapezius weakness is rare and is usually secondary to radical neck surgery in which the spinal accessory nerve is sacrificed.
Weakness in the scapular stabilizing muscles is another cause of scapular winging, and winging can also be associated with shoulder instability. Shoulder instability from recurrent shoulder dislocations can lead to weakness and dysfunction of the shoulder girdle musculature. Pseudowinging can be caused by an osteochondromata of the scapula.
The most common cause of scapular winging is injury to the long thoracic nerve that results in paralysis of the serratus anterior. This paralysis causes a dysfunction in rotation of the scapula. The primary mechanism seems to be acute or recurrent trauma during sporting events.
The long thoracic nerve seems to be susceptible to injury not only from trauma but also from viral illness sequelae, immunizations, and prolonged recumbency (as in prolonged surgical procedures under general anesthesia). It may also be injured through traction during shoulder movements, including shoulder depression or contralateral flexion of the cervical spine. Kauppila noted that the nerve was most mobile and prone to traction due to anterior movements of the scapula resulting in compression of the nerve by the inferior angle of the scapula. 
Diagnosis is based on history and physical examination. The patient usually complains of shoulder pain and loss of stabilization and weakness of the scapula in forward elevation. The deformity can be examined by having the patient push against a wall and abduct the arms over the head and looking for the asymmetry. EMG is useful in evaluating nerve damage.
Most atraumatic lesions will recover spontaneously. Conservative measures consist of bracing the scapula to the rib cage to help alleviate pain and stabilize the shoulder and to protect from overstretching the serratus anterior. It is important to avoid heavy lifting during the recovery phase. If there is no resolution and if pain and weakness are causing painful instability, surgical stabilization may be warranted.
Surgical stabilization includes pectoralis major tendon transfer to improve strength, with the goal of maintaining the inferior angle of the scapula against the chest wall while allowing the appropriate rotation for shoulder motion. Another procedure involves using a synthetic ribbon to loosely fix the inferior angle of the scapula to a rib below the level of the inferior angle to allow lateral movement of the scapula with elevation without winging and pain. The ultimate salvage procedures consist of scapuloplexy and scapulothoracic fusion.
Many patients who present with painful winging of the scapula find relief with conservative therapy; however, there exists a subset of this patient population in which such results, unfortunately, are not achieved. For such patients, the most common and successful surgical treatment for winging of the scapula, especially when it is caused by isolated palsy of the serratus anterior, is pectoralis major transfer.
In fact, pectoralis major transfer with autogenous soft-tissue augmentation has been described by several authors as the treatment of choice for scapular winging secondary to isolated serratus anterior palsy with injury to the long thoracic nerve.
Other surgical techniques (eg, simple bursectomy and bony fusion) have been recognized as treatment options for symptomatic scapular winging.  However, scapulothoracic arthrodesis is considered the definitive treatment for failure of previous surgical management. [20, 21] It is also thought to be the procedure of choice for patients with certain atraumatic causes of disabling scapular winging.
For patients with disabling pain and crepitation, failure of previous resection of the superomedial border of the scapula is an indication for scapulothoracic arthrodesis. For patients with intractable pain associated with fixed scapular winging or failed pectoralis transfer, indications for scapulothoracic arthrodesis include the following:
Difficulty reducing the scapula during a scapular stabilization test
Greater than 75% pain reduction during a scapular stabilization test
To perform the scapular stabilization test, the patient's scapula is stabilized to mimic the effect of a scapulothoracic arthrodesis. The patient is then asked to actively elevate the arm forward. With the scapula forcibly locked to the thoracic rib cage, the patient is asked to describe the amount of pain reduction achieved during active range of motion.
In a study by Krishnan et al, the scapulothoracic arthrodesis procedures were performed with a plate and wire fixation technique.  The plate and wire were positioned along the medial border of the scapula, and an autologous bone graft was used. Patients were then immobilized for 12 weeks, and rehabilitation was begun with gentle passive range of motion focusing on forward elevation and external rotation. After 6 weeks, progressive-resistance exercises were initiated.
Unfortunately, there were complications in more than half of the patients postoperatively, including pulmonary complications and pseudarthrosis.  Also, there were reports of persistent pain, but 91% of patients described improvement in their pain level and were pleased with the functional outcome. The high rate of patient satisfaction demonstrates that scapulothoracic arthrodesis may be a primary option or salvage procedure for patients with debilitating scapulothoracic dysfunction.
A study by Goel et al  was performed to investigate the clinical outcomes of patients who underwent scapulothoracic fusion (STF) to treat painful scapular winging. In this level of evidence IV retrospective review, 10 patients (12 shoulders) underwent STF and were followed for a mean period of 41 months (range, 8-72 months). Indications for STF included the following:
Three patients (five shoulders) with FHSMD
Four patients (four shoulders) with long thoracic or spinal accessory nerve palsy
Two patients (two shoulders) with painful loss of scapular control after excessive resection of the clavicle
One patient (one shoulder) with trapezius tendon dysfunction after previous spine surgery
Patients were not allowed to have shoulder motion for 4-6 weeks postoperatively; gentle pendulum exercises and passive range of motion were permitted. Active range of motion was permitted after 8-12 weeks.  CT was performed 6 months after surgery to check bony union of fusion. Preoperative subjective shoulder value averaged 39% and significantly improved to 63% postoperatively. There were also statistically significant improvements in range of motion (90º forward elevation preoperatively, 117º postoperatively), as well as pain.
Baseline (pain at rest) improved from 2.8 of 10 to 0.6 of 10 postoperatively.  Maximum visual analog scale scores (pain at its worst) improved from 7.8 of 10 preoperatively to 2.92 of 10 postoperatively. Despite these improvements, complications were common after STF, including pulmonary embolus (1/12), pleural effusions (4/12), hemopneumothorax (1/12), nonunions (2/12), and hardware removal (6/12).
Long thoracic nerve release is another surgical treatment for scapular winging due to serratus anterior muscle palsy.  This procedure is a relatively noninvasive and effective technique for this rare cause of scapular winging, usually due to a lesion in the nerve as a result of violent upper-limb stretching with compression of the nerve by the thoracodorsal artery.
A clinical study (level of evidence III) looked at eight patients who underwent long thoracic nerve release for scapular winging that failed conservative treatment and induced functional impairment (mainly pain).  Nine operations were performed in the eight patients. Neurolysis was performed along the long thoracic nerve, and collateral branches were spared. Postoperatively, patients were not immobilized, and they only used a sling to reduce pain. Activity was not restricted, but heavy lifting and sports were allowed on a case-by-case basis.
Follow-up was at 6 weeks and at 3, 6, and 12 months.  Neurologic recovery was checked with EMG at 9 months postoperatively. The patient with bilateral operations was lost to follow-up. Seven of the eight patients had statistically significant functional improvement, from a mean of 46.2 preoperatively to a mean of 66.7 at the end of follow-up. Pain was statistically significantly reduced at 6 weeks in all but one patient, from 5.8 preoperatively to 1.38 postoperatively; the last patient still had pain 3 months after the procedure.
At 6 weeks postoperatively, three patients had no clinical signs of scapular winging; by 6 months, all patients had no signs of scapular winging.  EMG was performed, and distal latencies of 8.4 ms preoperatively decreased to 5.5 ms postoperatively with at least partial serratus anterior reinnervation.
The authors concluded that neurolysis of the thoracic part of the long thoracic nerve to treat scapular winging secondary to serratus anterior muscle palsy seems to be a good first-line surgery, but because the pathology is so rare, it is difficult to conduct larger studies with greater statistical power. 
Scapulothoracic dissociation is a rare injury that is usually the result of severe trauma with lateral scapula displacement, clavicular disruption, and severe soft-tissue injury.  The displacement is the result of disruption of the muscular attachments of the scapula, and the scapula must become detached from the axial skeleton via clavicular fracture or sternoclavicular or acromioclavicular joint separation.
Traumatic scapulothoracic dissociation is associated with serious debilitating complications, the most important one being injury to the brachial plexus. Brachial plexus injuries usually are associated with a poor prognosis, which is attributed to the location of the lesion, commonly in the most proximal aspects. Associated subclavian or axillary artery vascular disruption can be life-threatening. [25, 26, 27, 28] There is one reported case of a patient who sustained scapulothoracic dissociation without neurovascular deficits. 
Diagnosis can be made on the basis of a nonrotated chest radiograph in which the scapula is laterally displaced. Lesions such as root avulsions and disruption of the cords have often been reported in the literature. Unfortunately, when the injury is described as a complete nerve injury by physical examination or when avulsion of the roots is confirmed by electrodiagnosis, there is little if any neurologic recovery. On the other hand, if the lesion is found to be incomplete, some neurologic recovery may be observed.
Patients with scapulothoracic dissociation require resuscitation and immediate surgical management to identify and repair the site of hemorrhage. Various techniques include repair of the arterial or vascular tree, subclavian artery or subclavian vein ligation, and, possibly, tamponade of collateral bleeding by packing of the dead space with laparotomy pads.
Unfortunately, the patient is left with a functionless or flail upper extremity as a result of proximal nerve damage to the cervical roots of the brachial plexus. The proximal nerve injuries do not regenerate and therefore have a poor prognosis for return of motor function. Patients are left with a nonfunctional upper extremity that is often a source of infection, unrecognized injury, and causalgia.
Rorabeck discovered that patients with complete brachial plexus injury were reported to return to work quicker if managed with early amputation, fitting of a prosthesis, and rehabilitation.  Consequently, it is recommended that patients with scapulothoracic dissociation undergo above-the-elbow amputation within 1 year of complete brachial plexus or cervical nerve root injury.
Above-the-elbow amputation may provide the best functional treatment option for flail extremity and a severe brachial plexus injury, such as complete brachial plexus avulsion. Reports of unsatisfactory outcomes during attempted repair of complete brachial plexus injuries have many authorities convinced that such lesions are not amendable to repair. It is accepted that above-the-elbow amputation can allow the patient to have a semiuseful upper extremity.
Facioscapulohumeral Muscular Dystrophy
FSHMD is an autosomal dominant myopathy that has complete penetrance but is variably expressed. The molecular defect is linked to chromosome 4q35 markers. Life span is usually normal. The incidence is 3-10 cases per million population, with an estimated prevalence of 1 in 20,000.
The age of onset is usually the second decade, and the presentation is one of facial and proximal weakness that progresses slowly. The weakness is usually in the face, shoulder girdle, and anterior portion of the legs. The proximal upper extremity muscles are usually weak, limiting the ability to carry heavy objects and raise objects above the shoulders. The biceps and triceps are usually weak with preservation of the deltoid and forearm and wrist flexors.
There are no electrocardiographic (ECG) or rhythm disturbances in FSHMD. The creatine kinase (CK) level is elevated to two to four times normal in about half of the patients. Early needle EMG studies may be normal in clinically involved muscles. As the disease progresses, EMG shows a myopathic picture with fibrillation and occasional repetitive discharges. The motor units become small and polyphasic, with an increased recruitment pattern. Muscle biopsy usually shows variability in sizes of fibers, with tiny fibers demonstrating a moth-eaten appearance.
Treatment focuses on activities of daily living. A supervised exercise program is recommended with slow, progressive increases in physical activity to prevent damage to the muscle or increase in weakness. As patients develop muscular weakness in the shoulder girdle, they may develop scapular instability secondary to weakness of the scapular stabilizers, limiting flexion and abduction of the arm. Scapulothoracic arthrodesis is the procedure of choice to improve upper-extremity function in patients with FSHMD. [32, 33, 34]
Berne et al described an initial average increase of 25º in abduction of the shoulder and 29º of forward elevation postoperatively.  All of the patients in this study had shoulder instability or weakness with scapular winging during active shoulder abduction.
Berne et al used the Horwitz maneuver (stabilization of the scapula against the thoracic wall by the examiner while the patient actively abducts the shoulder) as an indication for scapulothoracic arthrodesis.  A positive Horwitz maneuver was considered an indication for surgery, and a negative Horwitz maneuver contraindicated surgery. Good clinical results were achieved in more than 90% of patients. One major adverse side effect of the surgery is a possible decrease in vital capacity secondary to immobilization of the ribs.
In this study, postoperative exercises included isometric strengthening of the deltoid along with pendulum exercises.  Also, progressive-resistance training exercise was initiated to strengthen the shoulder muscles, with preoperative strength being achieved within 6 months. Many patients showed improved functional ability to perform activities of daily living and were able to continue performing their duties in the workplace.
In a retrospective study by Van Tongel et al,  medium- to long-term outcome of thoracoscapular arthrodesis with screw fixation for FSHMD was examined. The procedure was performed in 35 shoulders in 24 patients. The principal study group consisted of 21 patients (32 shoulders) who were followed for a minimum of 24 months.
Mean active arm elevation increased from 65º preoperatively to 119º postoperatively.  Pain improved from 9.8 out of 15 points preoperatively to 13.2 points postoperatively. Range of shoulder motion, strength, and patient satisfaction all improved postoperatively as well. There was no difference in functional outcomes in patients with follow-up of more than 10 years (eight shoulders) and follow-up of 2-5 years (12 shoulders).
Complications from surgery included a pneumothorax, superficial wound infection, and nonunion.  The fusion rate was 86%. The authors concluded that thoracoscapular arthrodesis using screw fixation improved both short-term and long-term shoulder function in patients with FSHMD.