➢ Three main theories explain the pathogenesis of the disabled throwing shoulder: the internal impingement theory, the posterior capsular contracture theory, and the scapulothoracic function theory.
➢ The throwing shoulder demonstrates a unique musculoskeletal profile of asymmetrically increased external rotation and decreased internal rotation with preservation of the total arc of motion (external rotation plus internal rotation) when compared with the contralateral shoulder. Most investigators agree that the total arc of motion should not exceed 186°. A 5° asymmetry in the total arc of motion can be predictive of increased injury risk.
➢ The three-dimensional motions of the scapula associated with arm motion during throwing demonstrate a consistent positioning of retraction, upward rotation, posterior tilt, and internal/external rotation for optimal glenohumeral stability. Discrepancies from this pattern have been associated with injury.
➢ Magnetic resonance imaging (MRI) scans with the shoulder positioned in abduction and external rotation can improve the accuracy of diagnosing SLAP (superior labrum, anterior to posterior) tears and partial-thickness rotator cuff tears.
➢ A minimalist approach should be taken when treating partial rotator cuff tears. Surgical repair should prevent tear progression while not limiting the substantial range of motion that is required for throwing.
Overhead throwing exposes the entire kinetic chain to excessive, cumulative stress, which ultimately can lead to injury. The repetitive and substantial angular velocities coupled with tremendous joint force can create instability, with ultimate tissue failure and resultant shoulder disability. It has been reported that shoulder torques generated by a professional baseball pitcher exceed the torsional failure limit in human cadaveric shoulders1. Over the past fifty years, there has been an increased understanding of the etiology of shoulder pain in the throwing athlete. In 1959, Bennett hypothesized that repetitive traction on the triceps insertion led to posterior capsular inflammation2. In 1977, Lombardo et al. noted excess fibrous tissue and ossification of the posterior capsule at the time of operative treatment in throwing athletes3. Several years later, Andrews et al. noted that pitchers were at risk for developing partial undersurface tears of the supraspinatus tendon4.
The dominant shoulder of a skeletally mature thrower demonstrates increased external rotation along with diminished internal rotation compared with the contralateral side5. In asymptomatic throwers, the total arc of motion is maintained but is shifted by 10° into external rotation6 as a result of bone and soft-tissue adaptations. During normal childhood development, humeral retroversion decreases from 78° in utero to 30° at skeletal maturity. However, repetitive throwing during growth restricts the derotation process7. Crockett et al. showed that professional baseball pitchers have 17° of increased humeral retroversion in the throwing shoulder as compared with the nondominant shoulder8. Furthermore, with throwing, the anterior capsule develops hyperlaxity, whereas the posterior capsule has the propensity to become thickened and contracted. Repetitive microtrauma to the anterior capsule during the late cocking phase of throwing leads to anterior laxity, articular-sided rotator cuff tears through impingement, or capsular injury through external rotation tensile overload. This critical moment in the throwing motion (the point of maximum external rotation) also has been shown to create a remodeling effect on the humerus as a result of repeated, torsional stress; this remodeling effect has been described as being both protective of the soft-tissue structures surrounding the glenohumeral joint (as a result of efficient energy transfer during throwing) and destructive to those structures9.
The term disabled throwing shoulder describes the limitations of function that prevent skeletally mature overhead athletes from playing at their maximum ability10. The causative factors, described by Burkhart et al.10, are divided into anatomical, physiological, and biomechanical categories. Anatomical factors include SLAP (superior labrum, anterior to posterior) tears, biceps tendon injuries, rotator cuff injuries, and instability. Physiological factors include muscle strength imbalance, muscle inflexibility, and glenohumeral internal rotation deficit. Biomechanical factors include kinetic chain deficits, scapular dyskinesis, and altered mechanics.
Pathophysiology of the Throwing Shoulder
There are three main theories regarding the development of the disabled throwing shoulder: the internal impingement theory, the posterior capsular contracture theory, and the scapulothoracic function theory. These theories correlate with the three groups of causative factors described by Burkhart et al.11. The internal impingement theory includes anatomical factors such as SLAP tears, biceps injuries, and rotator cuff injuries. The posterior capsular contracture theory encompasses glenohumeral internal rotation deficit with subsequent muscle strength imbalance. The scapulothoracic function theory encompasses biomechanical factors including kinetic chain deficits, scapular dyskinesis, and altered mechanics.
Internal Impingement Theory
Internal impingement, described as the entrapment of the rotator cuff between the humeral head and the glenoid labrum, may occur in either a posterosuperior or an anterosuperior location. Posterosuperior impingement occurs when the supraspinatus and infraspinatus tendons become entrapped between the greater tuberosity and the posterosuperior aspect of the glenoid labrum. This type of internal impingement is far more common and occurs with the arm in 90° of abduction and full external rotation. Although physiological contact between the posterosuperior aspect of the labrum and the rotator cuff does exist in asymptomatic individuals, repetitive motion often leads to pathological changes12,13. In addition, it has been shown that glenohumeral contact pressures substantially increase as the shoulder assumes higher angles of horizontal abduction while in the cocking position14. This extreme position increases the risk for partial rotator cuff tears and SLAP lesions13. In one series, thirty-six (100%) of thirty-six competitive athletes with partial-thickness rotator cuff tears had concomitant posterosuperior labral tears4.
Anterosuperior impingement is far less common than posterosuperior impingement and involves entrapment of the subscapularis tendon between the anterior aspects of the humeral head and glenoid labrum with the shoulder in flexion, adduction, and internal rotation15. This impingement leads to undersurface tearing of the subscapularis along with lesions of the long head of the biceps. It has been shown that as many as 50% of cases of biceps subluxation are associated with degenerative changes of the anterosuperior aspect of the glenoid labrum16.
Posterior Capsular Contracture Theory
Burkhart et al.11 noted that the posterior capsule must withstand tensile forces of up to 750 N during the deceleration and follow-through phases of throwing. The main restraints to this force are the infraspinatus tendon and the posterior band of the inferior glenohumeral ligament11. With repetitive eccentric contraction of the infraspinatus, the muscle and the posteroinferior capsule undergo hypertrophic changes and become stiff6. As a result of this posterior contracture, the center of rotation of the shoulder is shifted to a more posterosuperior location. This translation in the center of rotation in turn leads to posterosuperior instability when the shoulder is in an abducted and externally rotated position17. Secondary to this instability, external rotation of the humeral head is exaggerated, causing increased shear in the infraspinatus. Furthermore, a torsional force known as the peel-back phenomenon is applied to the biceps anchor during late cocking, contributing to the development of a SLAP lesion18.
Scapulothoracic Function Theory
The scapula plays a vital role in transferring energy from the trunk to the humerus during overhead throwing. The scapula of the throwing shoulder demonstrates increased abduction, protraction, and inferior translation19. During the late cocking phase of throwing, upward scapular rotation aids in maintaining glenohumeral joint symmetry20. However, weakness of the periscapular musculature can disrupt this relationship and alter the forces transmitted to the shoulder girdle. Altered force transmission leads to the subsequent development of the SICK scapula (i.e., scapular malposition, inferior medial border prominence, coracoid pain, and dyskinesis of scapular movement)10. Various pathological entities are associated with the SICK scapula: subacromial pain resulting from acromial malposition and decreased subacromial space, acromioclavicular joint pain resulting from anterior incongruity, superomedial scapular pain resulting from levator scapulae insertional tendinopathy, coracoid pain resulting from pectoralis minor tightness, and radicular pain resulting from thoracic outlet syndrome18. In addition to these inherent symptoms related to the abnormality itself, the SICK scapula also predisposes the shoulder to other injuries. Because the scapula is protracted and upwardly tilted, the glenoid faces anteriorly and superiorly, leading to anterior tension, posterior compression, and increased glenohumeral angulation. Posterior compression predisposes the shoulder to rotator cuff and labral tears, whereas increased glenohumeral angulation leads to increased external rotation, which exacerbates the aforementioned biceps peel-back effect.
Evaluation of the Thrower’s Shoulder
A wide spectrum of disorders of the kinetic chain, including issues related to the core, lower back, and hips, should prompt a complete examination of the throwing athlete. Initial shoulder symptoms often are related not to pain but rather to decreased control, decreased velocity, and difficulty warming up. Patients often are not able to recall a specific inciting event but rather note that the symptoms developed gradually over time. They also may present with symptoms similar to those of rotator cuff disease, such as a “dead arm” or shoulder weakness after throwing11. Patients also may complain of popping, snapping, and locking, which may occur in association with unstable labral tears. Athletes with posterosuperior impingement often complain of shoulder pain during the late cocking phase of throwing. Although the pain is typically localized to the posterior aspect of the shoulder, it also may be more generalized. To emphasize this point, Burkhart et al. highlighted that, in their series of ninety-six athletes with a disabled throwing shoulder, there was an 80% rate of anterior coracoid pain10.
A thorough and systematic physical examination of the thrower must include an assessment of the lower back, hips, and knees. The physical examination of the shoulder always begins visually, with observation of the patient’s posture and general appearance and how the patient is walking and moving the shoulder. Subsequently, with the patient facing away from the clinician, the clinician must look for any evidence of asymmetry between the two shoulders as well as for potential signs of atrophy and scapular winging. Scapular asymmetry can be measured by comparing the distance from the inferior angle of the scapula to the spinous process of the thoracic vertebra in three positions as described by Kibler21: with the arm at the side, with the humerus internally rotated and abducted 45°, and with the shoulder abducted to 90°. A side-to-side difference of >1.5 cm is deemed a positive lateral scapular slide test21. Tenderness to palpation should next be assessed along the glenohumeral joint line, the acromioclavicular joint, the long head of the biceps, and the coracoid process. Palpation of the biceps tendon is best performed with the arm in 10° of internal rotation because in this position the groove faces directly anteriorly22. Furthermore, tenderness over the coracoid suggests pectoralis minor tightness, which is often a result of scapular dyskinesis10.
Range of motion of both shoulders should be tested in both adduction and 90° of abduction. The thrower usually will have decreased internal rotation and increased external rotation, with maintenance of the total arc of motion (internal rotation plus external rotation) compared with the contralateral side; the total arc should average approximately 186°23. A total range of motion of >186° or a difference in range of motion of ≥5° when the throwing shoulder is compared with the contralateral shoulder can be predictive of the risk of injury in the throwing athlete. Strength testing should isolate the muscle being tested. The internal rotation lag sign is the most sensitive test of subscapularis function24, whereas supraspinatus strength can be assessed in the so-called empty can position25. Assessment of external rotation strength in adduction isolates the infraspinatus, whereas assessment of external rotation strength at 90° of abduction is used to test the teres minor.
As there is a strong association between internal impingement and SLAP lesions, clinical testing should coincide with these suspicions. Although there is general consensus in the literature that there is no one clinical test that can diagnose labral abnormality, the active compression test of O’Brien et al. has high sensitivity and specificity for SLAP lesions26. To perform the examination, the arm of the patient is forward elevated to 90°, adducted to 15°, and internally rotated. The patient is told to maintain this position while the examiner provides a downward force on the arm. Subsequently, the arm is brought into full external rotation while the other positions are maintained and a downward force is once again applied. The test is considered positive when the patient notes a greater amount of pain with the arm internally rotated along with a decrease in symptoms in full supination26. Last, specific scapular examinations include the scapular assistance test and the scapular retraction test. Both tests are important for diagnosing scapular incompetence by providing assistance with scapular motion. Scapular dyskinesis is present if manual stabilization increases strength in patients with apparent rotator cuff weakness and a protracted scapula6.
Imaging of the throwing shoulder is used as a confirmatory tool for pathological changes. The diagnosis of the painful throwing shoulder should begin with standard shoulder radiographs, including anteroposterior, axillary, and outlet views. Radiographs may reveal throwing shoulder-specific findings such as sclerosis and cystic changes of the greater tuberosity, excrescences of the inferior aspect of the glenoid, and impingement lesions of the posterior aspect of the glenoid24. Multiple studies have shown substantial changes on magnetic resonance imaging (MRI) scans of asymptomatic throwing shoulders27,28. MRI findings include partial-thickness rotator cuff tears, cystic changes of the greater tuberosity, and labral abnormalities, among other findings. Several additions to standard MRI of the throwing shoulder may be used to enhance diagnostic capabilities. MRI arthrography is routinely used to enhance the specificity, sensitivity, and accuracy of diagnosis. New evidence has revealed that the position of the arm can increase the diagnostic capabilities of contrast-enhanced MRI29. The authors29 found that an abduction and external rotation position of the arm can improve the accuracy of diagnosing anterior and posterior soft-tissue abnormalities, SLAP tears, and osseous defects of the glenoid. Lesniak et al. recently reported that MRI findings were correlated with the level of play but were not strongly predictive of disability or injury30.
Nonoperative Treatment of the Throwing Shoulder
Nonoperative treatment should follow a sequential, progressive, three-phased approach that highlights the entire kinetic chain while restoring glenohumeral joint mobility31. The first phase (the acute phase) should restore glenohumeral and scapulothoracic muscular activation by promoting synchronous muscle firing in which demands on the tissue are relatively low. Anti-inflammatory strategies are employed to prevent soft-tissue irritation while restoring the glenohumeral motion that is necessary for overhead throwing. The second phase (the recovery phase) emphasizes kinetic chain linkage through resistance training involving the core, lower extremity, and trunk. The third phase (the functional phase), imparts sport-specific, functional movement patterns necessary for a return to throwing. Plyometric exercise, shoulder end-range stabilization drills, and advanced isotonic strengthening are key components leading toward the initiation of a throwing program and eventual return to sport. Exercises during this phase are endurance-based, with a propensity toward high repetition and low resistance.
In an effort to prevent glenohumeral internal rotation deficit, the modified cross-body stretch and modified sleeper stretch should be utilized to improve internal rotation tightness and to maintain the total arc of motion bilaterally32. The clinical outcomes of nonoperative treatment have yet to be fully elucidated. In one of the few studies on this issue, Edwards et al. examined the role of nonoperative treatment specifically with regard to SLAP lesions33. The authors followed thirty-nine patients with known SLAP tears that had been diagnosed on the basis of a positive O’Brien test, pain in the bicipital groove, and MRI findings. Of the thirty-nine patients, fifteen were overhead athletes. Approximately 67% (ten) of the fifteen overhead athletes were able to return to throwing after being managed with a regimen involving nonsteroidal anti-inflammatory drugs (NSAIDs), scapular stabilization exercises, and posterior capsular stretching. Two recent articles highlighted the outcomes of nonoperative treatment in Major League Baseball players34,35. Of 115 players who were evaluated for a throwing injury, seventy-six were noted to have a SLAP lesion. In this subgroup, 41% (thirty-one) of seventy-six players were successfully managed nonoperatively, with 90% (twenty-eight) of the thirty-one being able to return to the previous level of play. In comparison, among throwers who underwent either SLAP debridement or repair, the rates of return to the previous level of play were 67% and 64%, respectively.
Operative Treatment of the Throwing Shoulder
Outcomes following operative intervention in competitive throwing athletes are often difficult to interpret because of the lack of standardized diagnosis, the duration of nonoperative treatment, the techniques used during operative treatment, and the postoperative rehabilitation protocol. Harris et al., in a recent systematic review, found that only approximately 68% of 287 elite throwing athletes were able to return to sport at approximately one year after surgery36. Although the players noted improvement in terms of pain and disability after surgery, the level of function did not return to preinjury levels. Another recent study of high-level athletes highlighted the difficulty of returning to sport after operative intervention35. That study specifically examined the outcomes for sixty-eight players with SLAP lesions, demonstrated with MRI and physical examination, who had had a failure of a primary attempt of guided physical therapy. The players were managed with a guided therapy algorithm focusing on correcting compensatory scapular dyskinesis, alleviating GIRD through posterior capsular stretching, and gradually returning to throwing through stepwise increases as demonstrated by pain-free throwing. Of the twenty-one pitchers who completed the nonoperative treatment algorithm, only 40% were able to return to play, with even fewer (22%) returning to play at the previous level of performance. Of the pitchers who underwent operative intervention, 48% were able to return to play and only 7% were able to return to play at the preinjury level of performance. Last, the article highlighted that nonpitchers had higher rates of return to play after both nonoperative and operative treatment. Among the position players who were managed with a course of nonoperative therapy, the rate of return to play was 39% and the rate of return to the preinjury level of performance was about 26%. Among the nonpitchers who were managed with operative intervention, the rate of return to play was 85% and the rate of return to the preinjury level of performance was 54%.
Although operative treatment leads to symptomatic improvement after surgery, the prognosis in terms of returning to asymptomatic throwing and to the previous level of performance is guarded. For this reason, it is our recommendation that the throwing athlete undergo an initial guided therapy regimen before operative intervention. Because of the normal adaptive changes of the throwing shoulder, operative intervention should be considered with a so-called less-is-more approach. Any operative intervention should be aimed at recreating the anatomy inherent to the throwing shoulder, and, in cases of intraoperative decision-making, a minimalist approach often provides superior outcomes over aggressive surgical intervention.
SLAP Tears and the Biceps Tendon in Throwing Athletes
Surgical intervention for SLAP tears in throwing athletes presents a challenge to the orthopaedic surgeon. Several series have highlighted the fact that there is much variability in the outcomes of surgical repair in throwing athletes, with the rate of successful outcomes ranging from 22% to 94%34,37-40. Unfortunately, there is no accepted standard treatment algorithm that includes the time to operative intervention, the location of the SLAP tear, the presence of abnormalities involving additional sites, the operative technique, or the postoperative rehabilitation protocol. Furthermore, the intraoperative assessment of what constitutes a SLAP tear is subject to controversy41.
Recent data have suggested that some factors can have an impact on the outcomes of surgical repair of SLAP lesions. Provencher et al. examined 179 patients who had undergone surgical repair of a type-II SLAP tear42. Of the 179 patients who were available for follow-up, 36.8% met the criteria for “failure” of operative repair, with an age of more than thirty-six years being cited as the most important prognostic factor for failure following operative intervention. Park et al., in a recent study of the outcomes of operative repair for type-II SLAP tears in elite overhead athletes43, found that subjective scores improved after operative intervention but that there was a nonsignificant trend toward diminished return to play, with only 38% of elite baseball players returning to play as compared with 75% of all overhead athletes. Risk factors regarding revision surgery following SLAP repair have been studied. Frank et al. found that an age of less than twenty years and participation in sports as an overhead thrower increased the risk of revision surgery after the repair of a SLAP tear44. These findings, in addition to the presence of a coexisting rotator cuff tear, have negative prognostic indications for patients undergoing operative repair of a SLAP tear45-48.
McCormick et al. recently highlighted the role of addressing the pathomechanical function of the biceps during revision SLAP repair49. Forty-six patients who had had a failure of arthroscopic repair were managed with open subpectoral biceps tenodesis. After an average duration of follow-up of 3.5 years, the patients had substantial improvement in both subjective and objective scores, with an ultimate rate of return to active military duty and sports of 81%. The findings of that study, in addition to previous findings on the relatively good outcomes following biceps tenodesis for the treatment of SLAP tears in patients with a high risk for failure, lend credence to consideration of biceps tenodesis for the treatment of SLAP tears in a certain subset of patients50. These findings were supported by Denard et al., who reported that biceps tenodesis provided a faster recovery, more predictable outcomes, and a faster return to activity in comparison with SLAP repair alone when used for the treatment of type-II SLAP tears in nonthrowing athletes older than thirty-five years of age51.
Several recent cadaveric studies have also examined the importance of the biceps with respect to SLAP tears. Strauss et al., in a recent cadaveric study, found that biceps tenodesis does not alter the translational kinematics of a superior labrum-deficient shoulder52. The authors found that repair of anterior superior labral lesions did not restore anterior translation to the baseline state but that it did improve posterior and combined abduction-external rotation translation. The addition of a biceps tenodesis did not significantly alter the translational properties in the superior labral tear model or undermine the restored kinematics in the repaired model. For unstable posterior SLAP tears that require repair, we routinely use knotless suture anchors (PushLock; Arthrex, Naples, Florida) (Fig. 1). The use of knotless anchors has been advocated by some surgeons in an effort to provide a low-profile implant that minimizes abrading from the suture material53.
Rotator Cuff Injury in Throwing Athletes
Consensus with respect to the treatment of rotator cuff tears in throwing athletes is lacking because of the intricacy of the pathoanatomy of the thrower’s shoulder, the high level of demand for the extremity, and the use of nonstandardized surgical techniques and treatment principles. The three options for the surgical treatment of rotator cuff tears in throwing athletes include debridement, repair of delamination with suture, and repair of the tendon tear to bone. As previously mentioned, SLAP tears have a more guarded prognosis following surgical repair in comparison with rotator cuff tears47. In a recent study, Van Kleunen et al. retrospectively reviewed the records for seventeen high-level throwing athletes who had undergone concomitant surgical repair of a SLAP tear with standard anchor placement and surgical treatment of a partial-thickness infraspinatus tear with either anchor placement or intralaminar repair54. The authors found that players had a poor rate of return to play, with only 35% (six) of seventeen patients returning to play at a level that was similar to the preinjury level. The authors also found that there was a substantially lower level of return to play and also lower KJOC (Kerlan-Jobe Orthopaedic Clinic) shoulder and elbow scores for patients who underwent repair of the infraspinatus tear with suture anchor placement as opposed to intralaminar repair. Ide et al. reported similar findings in a study of patients who had a partial-thickness tear that was repaired to bone, with approximately one-third of overhead throwing athletes returning to sports at the preinjury level55. This finding confirmed prior knowledge that repairing the tendon to bone may tether the compensatory anatomy of the rotator cuff and lead to less-than-ideal outcomes56,57. We contend that the surgeon should take a minimalist approach to dealing with partial rotator cuff tears and should take steps to perform a surgical repair that prevents progression of the tear but does not limit the substantial range of motion that is required for competitive play. For this reason, we typically employ either debridement or intratendinous repair of delamination tears as opposed to a repair to bone with a suture anchor (Fig. 2). Although this recommendation is based on limited evidence, such an approach has been recommended in several recent review articles58,59.
Injuries involving the shoulder in skeletally mature throwing athletes represent a unique treatment challenge to many orthopaedic providers. Appropriate treatment is informed by a thorough understanding of the pathomechanics of the thrower’s shoulder as well as a treatment regimen that is aimed at returning the athlete back to peak performance.
Source of Funding: No external funds were received in support of this study.
Investigation performed at the Center for Shoulder, Elbow, and Sports Medicine, Department of Orthopedic Surgery, Columbia University Medical Center, New York, NY
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, no author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
- Copyright © 2015 by The Journal of Bone and Joint Surgery, Incorporated