Shoulder and Elbow Injuries in Adult Overhead Throwers: Imaging Review
Abstract
Overhead throwing, particularly in baseball, subjects the shoulder and elbow to various unique injuries. Capsular contracture following repetitive external rotation shifts the humeral head posterosuperiorly, predisposing to glenohumeral internal rotation deficit (GIRD), Bennett, posterosuperior internal impingement (PSI), and superior labrum anterior-posterior (SLAP) lesions. GIRD represents loss of internal rotation at the expense of external rotation. Bennett lesion represents ossification of the posteroinferior glenohumeral ligament due to repetitive traction. PSI manifests with humeral head cysts and “kissing” tears of the posterosuperior cuff and labrum. Scapular dysfunction contributes to symptoms of PSI and predisposes to labral or rotator cuff disease. “Peel-back” or SLAP lesions occur when torsional forces detach the biceps-labral anchor from the glenoid. Finally, disorders of the anterior capsule, latissimus dorsi, teres major, and subscapularis are well recognized in overhead throwers. At the elbow, injuries typically involve the medial-sided structures. The ulnar collateral ligament (UCL) is the primary static restraint to valgus stress and can be thickened, attenuated, ossified, and/or partially or completely torn. Medial epicondylitis can occur with tendinosis, partial tear, or complete rupture of the flexor-pronator mass and can accompany UCL tears and ulnar neuropathy. Posteromedial impingement (PMI) and valgus extension overload syndrome are related entities that follow abundant valgus forces during late cocking or acceleration, and deceleration. These valgus stresses wedge the olecranon into the olecranon fossa, leading to PMI, osteophytes, and intra-articular bodies. Other osseous manifestations include olecranon stress fracture and cortical thickening of the humeral shaft.
©RSNA, 2023
Quiz questions for this article are available in the supplemental material.
Introduction
Overhead throwing injuries are common with many sports, most frequently impacting the shoulder and elbow. A study of 112 collegiate athletes found shoulder injuries to be common with baseball (41 cases), swimming (24 cases), softball (17 cases), tennis (15 cases), and volleyball (15 cases) (1). Overhead motion during a volleyball spike, tennis serve, or swimming stroke can predispose an individual to injury (1,2), although the baseball pitch is by far the most commonly used model. The abundant forces and rapid shifts in joint position from the overhead pitching motion place the shoulder and elbow stabilizers at considerable risk for multiple unique injuries. In a study of North American professional baseball players from 2011 to 2016, 3090 injuries were recorded, with 511 (17%) of them affecting the shoulder, making it the most vulnerable joint (3). Amongst player positions, the pitcher was the most prone to injury, occurring in 78% of cases (3). In terms of elbow disorders, data from the Major League Baseball Health and Injury Tracking System (4) revealed 3185 elbow injuries (430 in major league, 2755 in minor league) during the 2011–2014 seasons. Pitchers were also the most likely players to have elbow injuries and require surgery, and they had the greatest mean number of days missed from play when they were treated nonsurgically.
Throwing Motion Overview
Overhead throwing in baseball involves one continuous motion (ie, kinetic chain) that can be divided into six phases: windup, stride, cocking, acceleration, deceleration, and follow-through (Fig 1). These six phases are relatively consistent in terms of overhead throwing activities and can vary slightly on the basis of competition level and pitching or throwing style. The overall motion to propel a ball forward at high speeds requires coordinated efforts from the proximal joints (torso and lower extremities) and distal (arm) segments (5). Shoulder and elbow movements are emphasized in this review.
Figure 1. Graphic illustration of the six phases of throwing: windup, stride, cocking, acceleration, deceleration, and follow-through. Injuries of the shoulder are common during all phases except windup and stride, whereas injuries of the throwing elbow transpire mainly during late cocking and acceleration, when maximal valgus forces are generated. ER = external rotation, IR = internal rotation.
Throwing begins with windup, which involves shifting the body’s overall center of mass from both legs to a single-leg stance. This phase places minimal stress on the shoulder and concludes with the joint in slight internal rotation and abduction (2). The elbow is flexed throughout this phase and, like the shoulder, is subjected to minor muscle activity (6). Stride begins with separation of the arms to nearly 180° apart and ends at foot strike (6,7). The shoulder begins external rotation and horizontal abduction, while the elbow extensor and flexor muscles are activated to control elbow motion. The elbow is extended during the first half of this stage and flexed during the second half. At foot strike, the elbow is flexed between 80° and 100° (6). Similar to windup, stride requires negligible elbow kinetics and muscle activity.
Arm cocking starts with foot strike and concludes with the shoulder in maximal external rotation. While the proximal muscles (upper trunk and pelvis) are rotating and extending, the distal muscles are externally rotating and abducting (shoulder), or flexing (elbow) (5). Scapular elevation and upward rotation help to ensure that there is sufficient subacromial space to accommodate the 80°–100° of humeral abduction and thus avoid impingement (5). As the shoulder externally rotates the arm, valgus forces of up to 64 N are produced at the elbow (8). This valgus load subjects the medial elbow structures, especially the ulnar collateral ligament (UCL), to large tensile forces. Maximal abduction and external rotation (ABER) of the shoulder and valgus torque on the elbow mark the conclusion of this phase and are often termed the critical moment, as they are often implicated in both adaptive and pathologic changes in both joints (5). This movement is also referred to as the external rotation set point, or “slot,” which is critical in maximizing pitching velocity (9).
Acceleration is the time between maximal external rotation and ball release. As the shoulder moves rapidly from external rotation into horizontal adduction and internal rotation, the elbow undergoes rapid extension from 120° of flexion to nearly 25° of flexion at ball release (6). These rapid shifts in movements of both joints create the shortest but most intense phase of the pitching motion (2).
Deceleration, which follows acceleration, is considered the most violent phase. It begins at ball release and ends at maximal shoulder internal rotation and elbow extension. Glenohumeral joint loading is greatest during this phase, as the shoulder attempts to safely slow down forward arm progression (7). The scapula protracts, providing a stable base as the humerus rapidly switches from external to internal rotation. The biceps, brachialis, and elbow flexors prevent rapid elbow extension and forearm pronation to avoid the olecranon from impinging on the olecranon fossa (10). An eccentric elbow flexion torque of approximately 10–35 N ⋅ m is generated throughout the deceleration phase to slow down elbow extension (6).
Follow-through commences with the shoulder in maximal internal rotation and ends when the thrower reaches a balanced position. Movements of the larger muscles in the trunk and lower extremities dissipate energy in the throwing arm (7). The phase concludes with the shoulder in internal rotation and adduction while the elbow flexes into a comfortable position as the trunk rotates forward and the arm moves across the body (6).
To maximize the ball velocity, the elite thrower must maximize internal rotation velocity of the arm. Depending on the pitch type, elite baseball pitchers can reach angular velocities of 7000°–7900° per second (2,9) by means of extreme external rotation in late cocking, thereby maximizing the rotation arc. Burkhart and colleagues (9) concluded that repetitive posteroinferior capsular contracture from the follow-through phase is the first event in the “disease cascade” in the throwing shoulder. During the cocking phase, the contracted posteroinferior glenohumeral ligament translates below the humeral head, acting as a lever and forcing the humeral head posterosuperiorly. This shift in contact point allows clearance of the greater tuberosity to achieve maximal external rotation. Elite pitchers recognize this “set point” to deliver effective pitching velocities (9).
Throwing Shoulder Injuries
Glenohumeral Internal Rotation Deficit
In the throwing arm, glenohumeral internal rotation deficit (GIRD) represents internal rotation loss to achieve maximal external rotation (9).
Loss of internal rotation may be adaptive and asymptomatic initially if the total arc of motion remains symmetric between the shoulders. Clinically relevant GIRD may occur when the internal rotation loss exceeds the external rotation gain, creating a deficit in the total arc of motion
Figure 2. Clinical photographs demonstrate preservation of the normal (180°) arc of motion in the nonthrowing arm (A), versus adaptive external rotation in the throwing arm (B), versus GIRD in the throwing arm (C). In a thrower’s arm, the gain in external rotation is balanced by the loss of internal rotation, preserving the total arc of motion (B). With GIRD (C), a pathologic loss of internal rotation results in a loss of the total arc of motion of 5° or greater (comparing A vs C) and an absolute loss of internal rotation of 20° or greater (comparing B vs C).
The current threshold for total arc of motion deficit is a 5° or greater difference between the shoulders (11). The most accepted threshold for relative loss of internal rotation compared to that of the contralateral shoulder is greater than or equal to 20°, with symptoms occurring beyond 25° (9). Glenohumeral rotation is measured by using a goniometer, with the shoulder abducted 90° and the scapula stabilized against the examination table or with the patient sitting (12).
A widely accepted cause of GIRD is a severely contracted and thickened posteroinferior recess and capsule (9). Evidence supporting this model includes enhanced internal rotation after selective capsulotomy (9) and clinical improvement following posterior capsular stretching (13). More recently, other authors (11) have suggested humeral retroversion and muscular stiffness due to repetitive strain as additional important contributors.
The posteroinferior glenohumeral ligament and axillary recess capsular thickening are well depicted on axial and coronal oblique MR images (Fig 3) (14). At arthroscopy, a posteroinferior glenohumeral ligament thicker than 6 mm has been suggested (9), although to date, no available reference value for posterior capsular thickness exists. An MR arthrography study by Tuite et al (15) of 26 overhead throwing athletes with pathologic posterosuperior internal impingement (PSI) and GIRD showed that throwers tend to have a thicker labrum and posteroinferior capsule and a more shallow posterior capsule-recess angle than do nonthrower controls (Fig S1) (15).
Figure 3. Posteroinferior glenohumeral ligament and capsule thickening in a 20-year-old former high school baseball pitcher with clinical and physical examination findings of GIRD. Axial (A) and sagittal (B) T2-weighted fat-suppressed MR images show thickening of the posteroinferior glenohumeral ligament and capsule (arrows) adjacent to the posteroinferior labrum (arrowheads).
Bennett Lesion
A Bennett lesion, or thrower’s exostosis, is an extra-articular ossification along the posteroinferior glenoid rim. While this condition is commonly asymptomatic, the lesion may be painful owing to nonunion of fragments, acute avulsion of the ossification, or capsular and axillary nerve irritation (16). Infraspinatus muscle atrophy from suprascapular nerve impingement by the exostosis also has been reported in athletes who play overhead sports (Fig 4), especially pitchers and volleyball players (17). In pitchers, infraspinatus muscle atrophy may occur in the absence of the ossification and has been reported in up to 4% of major league starting pitchers (18). Cummins and colleagues (18) hypothesize that this is related to cumulative neuropraxia on the suprascapular nerve at the spinoglenoid notch, even in the absence of ossification.
Figure 4. Bennett lesion in a 45-year-old retired professional baseball pitcher with a history of long-standing right shoulder pain. (A) Axillary radiograph shows a bony exostosis (arrowhead) in the posterior glenoid, consistent with a Bennett lesion. (B) Findings on the corresponding axial T1-weighted MR arthrogram confirm the exostosis (arrowhead) and complete fatty atrophy of the infraspinatus muscle (*).
The pathophysiology of a Bennett lesion is still being debated. Hypotheses include traction of the posteroinferior glenohumeral ligament during the deceleration phase and humeral head impingement on the posterior glenoid during the cocking phase (19). Although the lesion may be a marker for repetitive capsular traction in throwers, there is no consensus as to whether it is truly pathologic. Wright and Paletta (20) reported a Bennett lesion in 22% of 55 asymptomatic major league pitchers and found no correlation between the presence of the lesion and thrower age, years pitched, or innings pitched. Conversely, a more recent study (19) found that patients with Bennett lesions had played baseball longer (10.6 years) than those without Bennett lesions (8.8 years).
The association between the exostosis and other PSI findings, including undersurface rotator cuff tears and posterior labral tears, is also controversial (16,19). Park et al (19) contend that the lesion tends to form parallel to and not toward the inner aspect of the glenoid and emphasize that patients experience pain mostly during follow-through rather than during cocking. While the exostosis may represent an advanced form of the posterior capsular thickening seen in GIRD, there is no definite established relationship. An investigation involving 388 baseball players found no significant difference in internal rotation deficit between patients with and those without Bennett lesions (19). Given variable evidence in the literature to support an association with signs and symptoms of PSI and GIRD, the finding of a Bennett lesion must be correlated with the clinical history and physical examination findings.
The ossification can be visualized on axillary, Stryker notch, or modified Bennett radiographs. CT can be used to assess for fragmented lesions or posterior glenoid dysplasia (19). Sagittal and axial T2-weighted or proton-density–weighted MR images obtained parallel to the glenoid fossa depict the Bennett lesion as a low-signal-intensity crescentic bump with an elevated periosteum along the posteroinferior glenoid (19). While CT remains the reference standard for diagnosing Bennett lesions with a sensitivity and specificity of 98% and 97%, respectively, T1-weighted MR images can depict early thickened subperiosteal tissue before mature mineralization (19).
Nakagawa et al (21) proposed four criteria for diagnosing a symptomatic Bennett lesion: (a) posterior glenoid rim bony spur on radiographs; (b) posterior shoulder pain while throwing, particularly during follow-through; (c) posteroinferior glenohumeral joint tenderness; and (d) pain improvement following local anesthetic injection. For pain that is resistant to conservative treatment for longer than 3 months, the authors recommend arthroscopic lesion removal via Bennett-plasty (16,21). Meister (2) showed successful return to play in 10 of 18 players when resection, along with débridement of the rotator cuff and labral tears, was performed. The investigation of Yoneda et al (16), involving baseball players with concomitant rotator cuff and labral disease, concluded with 11 of 16 players resuming competitive activity.
Posterosuperior Internal Impingement
PSI refers to abnormal contact of the undersurface of the posterior supraspinatus and/or anterior infraspinatus tendons between the greater tuberosity and the posterosuperior glenoid and labrum during late cocking (Fig 5). Walch et al (22) first attributed pathologic and symptomatic impingement, initially considered to be physiologic and asymptomatic in throwers, to repetitive microtrauma from the throwing motion. The clinical diagnosis is established by the presence of pain without apprehension in the ABER position and relief on application of a posteriorly directed force (Jobe relocation test), eliminating the impingement (22).
Figure 5. Graphic illustrations of PSI show abnormal contact of the undersurface of the posterior rotator cuff (specifically, the posterior supraspinatus and anterior infraspinatus) between the greater tuberosity and the posterosuperior glenoid and labrum (between arrowheads in B) during late cocking (arrow). The biceps tendon (highlighted in blue in B) is shown as a landmark.
Various theories regarding the causes of PSI exist. Several investigators (9,11,12,15) believe it represents a consequence of altered biomechanics from a stiff posteroinferior capsule (ie, GIRD). Myers et al (12) showed significantly greater GIRD and posterior shoulder tightness in 11 throwing athletes with impingement compared to controls, with an average internal rotation loss of 19.7°. Others postulate that there is an association of impingement with anterior instability. The excessive external rotation during throwing elongates the inferior glenohumeral ligament complex, allowing significant increases in anterior translation (23). The consequent anterior capsular laxity and microinstability exacerbate the entrapment of the posterosuperior labrum and capsule between the humeral head and glenoid (23).
The posterosuperior shift in contact point between the glenoid and the humeral head may contribute to pathologic shearing forces on the rotator cuff and labrum predisposing to tears. MRI features include humeral head cysts, articular-sided partial-thickness tears of the posterior supraspinatus and anterior infraspinatus tendons (Fig 6) and posterosuperior labrum tear
Figure 6. Humeral head cysts and an undersurface rotator cuff tear in a 26-year-old female elite volleyball player, with concern for a labral tear. Coronal (A) and sagittal (B) T2-weighted fat-suppressed MR images show fluid signal intensity on the undersurface of the infraspinatus tendon at the footprint, consistent with a partial-thickness tear (between the white bars). The overlying bursal-sided fibers (arrowheads in A) are intact. There are also a few cysts with surrounding edema-like changes (arrows) in the osseous footprint of the infraspinatus tendon; these cysts are more optimally visualized in B. These findings are commonly seen with PSI.
Cysts in the absence of clinical or other MRI features of impingement can be mistaken for marrow and cortical abnormalities of Hill-Sachs lesions and should be carefully scrutinized on sequential axial images. While the cortical irregularities, defects, and/or cysts of PSI are seen in the posterior humeral head, the Hill-Sachs lesion affects the posterosuperior humeral head—specifically the three superiormost axial images (26) or the superior 4–5 mm of the humeral head (Fig 7) (7,27). Caution must also be exercised with articular-sided partial-thickness tears presenting in asymptomatic throwers (25,28). MRI studies of 20 asymptomatic shoulders in elite overhead athletes showed rotator cuff tears in up to 40% of the players (29). MR arthrography with ABER positioning is especially useful for confirming undersurface cuff tears by lifting the tendon from the humeral head (Figs 8, S2) (24).
Figure 7. Hill-Sachs lesion versus PSI in two patients. (A, B) Axial T2-weighted fat-suppressed (A) and coronal T1-weighted (B) MR images in a 34-year-old woman who sustained a shoulder dislocation a week earlier show a prominent bony concavity (arrows) on the posterosuperior humeral head, at the top or superior aspect of the humeral head, consistent with a Hill-Sachs lesion. (C, D) Axial T2-weighted fat-suppressed (C) and coronal T1-weighted (D) MR images in a 36-year-old right-handed baseball pitcher without any history of instability show bony concavities (arrowheads) in a lower axial section at the level of the glenoid. These findings favor sequelae of PSI.
Figure 8. Undersurface rotator cuff tear in a 28-year-old male elite beach volleyball player with persistent shoulder discomfort. T1-weighted fat-suppressed ABER MR arthrogram shows contrast imbibition through the torn undersurface of the infraspinatus tendon (arrow), undercutting the detached posterior labrum (arrowhead).
Scapulothoracic Muscle Dysfunction
Scapulothoracic muscle dysfunction, or SICK (scapular malposition, inferior medial border prominence, coracoid pain and malposition, and dyskinesis of scapular movement) scapula syndrome, is a fatigue condition that disrupts the normal humeral and scapular alignment during acceleration, contributing to symptoms of PSI (23). The characteristic clinical manifestation is asymmetric malposition of the scapula in the dominant throwing shoulder; this frequently appears with one shoulder lower than the other (30). Types I (inferior medial scapular border prominence) and II (medial scapular border prominence) scapulothoracic muscle dysfunction are associated with labral disease, and type III (superomedial border prominence) is associated with rotator cuff lesions (30). In a cadaveric study (31) involving simulation of the throwing motion, greater internal scapular rotation and less upward scapular rotation significantly increased the glenohumeral contact pressure and impingement of the rotator cuff tendon between the greater tuberosity and the glenoid.
SLAP Tears and Pseudolaxity
The shift in glenohumeral contact point predisposes an individual to labral tearing, specifically the “peel-back” injury of the biceps-labral complex. In the ABER position, the posteroinferior glenohumeral ligament is positioned beneath the humeral head. A contracted posteroinferior capsule levers and pushes the humeral head posterosuperiorly (Fig 9). In this position, the biceps tendon twists upon its base, forcing the posterosuperior labrum and the biceps root to rotate medial to the glenoid, predisposing to biceps anchor and labral detachment from bone (Fig 10).
The associated labral injury is a SLAP tear, classified as type II (Fig 11) (32). While GIRD is considered the remote or chronic cause of dead arm, a SLAP II tear is considered the immediate or acute cause
Figure 9. Drawing superimposed on a sagittal T1-weighted MR arthrogram shows the shift in position that occurs in the major tendon and capsuloligamentous structures of the glenohumeral joint between the resting position (all graphics in yellow) and ABER position (all graphics in blue). AIGHL = anterior band of the inferior glenohumeral ligament, MGHL = middle glenohumeral ligament, PIGHL = posterior band of the inferior glenohumeral ligament, SGHL = superior glenohumeral ligament.
Figure 10. Graphic illustrations demonstrate the mechanism of peel-back injury. The normal position of the biceps tendon is highlighted in blue. In the ABER position, the new position of the humerus forces the biceps tendon to twist (black arrow) and change position (* in A), stripping the biceps anchor and posterosuperior labrum (white arrow in B) from the osseous attachment.
Figure 11. Peel-back lesion in a 28-year-old male elite beach volleyball player (same patient as in Fig 8). (A, B) Axial (A) and ABER (B) T1-weighted fat-suppressed MR arthrograms show contrast material extending through the superior labrum, from the anterior to posterior aspects (arrowheads), consistent with a thrower’s SLAP or peel-back lesion. (C) Subsequent arthroscopic findings confirmed the tear (arrowheads).
The peel-back injury may be visualized in the ABER position (Fig 11) with MR arthrography (34). The following grades have been proposed for assessment of the posterosuperior labral position relative to the glenoid articular plane: grade 0, whereby the labrum is lateral and cranial to the glenoid tangent line; grade 1, whereby the labrum remains flush but not medial and caudal to the glenoid tangent line; and grade 2, whereby the labral apex is clearly positioned medial and caudal to the glenoid tangent line (34).
In the ABER position, the anteroinferior capsule is taut across the humeral articular surface; this is the so-called cam effect (9). According to Burkhart’s circle concept, stresses on the posterosuperior labrum allow laxity to be channeled on the opposite side of the labral ring or anteroinferior labrum (9). In throwers, the shift in glenohumeral contact diminishes the cam or tightening effect of the anteroinferior capsule, resulting in capsular redundancy and thereby permitting greater external rotation (Fig 12) (9).
Figure 12. Graphic illustration of anterior pseudolaxity. When the shoulder is in the ABER position, the anteroinferior capsule is tightly draped across the inferior articular surface of the humerus (curved arrows); this condition is also termed the cam effect against the proximal humerus. With the posterosuperior shift of the humeral head on the glenoid (straight arrow), the cam effect is reduced, leading to anterior capsular redundancy or pseudolaxity (arrowheads).
Anterior Shoulder Injuries
In professional baseball pitchers, subscapularis tears tend to occur within the inferior half of the muscle, instead of the more common superior insertional tendon tears (Fig 13) (7,35). Excessive external rotation and cross-body abduction during late cocking stretch the inferior fibers across the anterior glenohumeral joint to their limit, predisposing to subscapularis muscle failure. As the humerus continues to externally rotate between late cocking to early acceleration, the subscapularis begins internal rotation and forward acceleration (7). Lesser tuberosity avulsions also may occur but are more common in adolescent throwers (36). Avulsions are best depicted with axillary radiography or CT, while MRI can be used to evaluate the integrity of the subscapularis, degree of retraction, and presence of muscle atrophy (7).
Figure 13. Subscapularis tear in a 24-year-old male professional pitcher with soreness in the shoulder for 1 week and a previous teres major strain. Sagittal (A) and axial (B) T2-weighted fat-suppressed MR images show a high-grade injury of the subscapularis muscle involving the lower myotendinous units (straight arrows in A). The uppermost portion of the subscapularis muscle is relatively spared (curved arrow in A). There is fluid surrounding the injured myotendinous units and the latissimus dorsi–teres major muscle-tendon complex (arrowheads in A). There is marked retraction of the torn muscle bellies (arrows in B).
Latissimus Dorsi and Teres Major Injuries
Latissimus dorsi and teres major injuries occur during late cocking and acceleration. The latissimus dorsi muscle adducts the elevated arm against resistance to pull down on the humerus and compresses the inferior scapula when the arm is elevated (37). The teres major muscle acts to adduct, extend, and internally rotate the humerus. Injuries to these muscles can occur together by virtue of their confluent humeral insertion, synergistic muscle action, and similar muscle vectors (37). In a study of 12 cadaveric specimens, all shoulder specimens demonstrated fascial connections between the two muscle bellies, while the tendon insertions were most commonly separate in two-thirds of specimens, or loosely bound or completely joined in the remaining one-third (38). Most acute injuries involving one or both tendons have been reported in professional baseball pitchers (Fig 14) (37) and less commonly in basketball and volleyball players (39).
Figure 14. Normal latissimus dorsi (LD) and teres major (TMa) anatomy versus latissimus dorsi and teres major injuries. CB = coracobrachialis muscle, LH-tr = long head of the triceps muscle, SH-b = short head of the biceps muscle, SSc = subscapularis muscle, TMi = teres minor muscle. (A, B) Axial (A) and sagittal (B) proton-density–weighted fat-suppressed (PDFS) MR images in an asymptomatic healthy volunteer demonstrate the normal anatomy of the latissimus dorsi and teres major in comparison to the injured patient. (C, D) Axial (C) and sagittal (D) PDFS MR images in a 28-year-old professional baseball player show a moderate-grade partial-thickness tear involving the distal insertion of the combined tendon of the latissimus dorsi and teres major (arrowheads in A, C, and D). Arrow in A and C points to the pectoralis major tendon.
The diagnosis may be challenging in the setting of nonspecific pain in the axilla at ball release and pain with resisted pull-downs. The clinical history must be considered, as standard shoulder MRI protocols do not routinely include the latissimus dorsi, which requires larger field-of-view T2-weighted fat-suppressed sequences for assessment (37).
Throwing Elbow Injuries
Abundant valgus loads during acceleration generate large tensile forces at the medial elbow, resulting in ligament tears, tendon injuries, ulnar nerve damage, and osseous abnormalities (6).
UCL Injuries
The UCL is the primary static restraint to valgus stress on the elbow and thereby the most commonly injured structure in overhead throwers.The anterior bundle is further subdivided into anterior and posterior bands, which exhibit different strain and injury patterns depending on the elbow position (41). The anterior band exhibits an isometric strain pattern throughout the elbow range of motion, while the posterior band strain pattern increases at higher degrees of elbow flexion. The posterior bundle contributes to stability at flexion angles greater than 120°, and the transverse bundle provides minimal stability to valgus forces at all angles (Fig 15) (42,43).
Figure 15. Graphic illustration of the bundles and bands of the UCL: the anterior bundle (yellow), transverse bundle (red), posterior bundle (purple), and annular ligament (AL).
An abnormal UCL can be thickened, attenuated, ossified, and/or partially or completely torn. Most injuries involve the anterior bundle and occur in its midsubstance but can involve other locations or more than one location (40,41). The anterior band of the anterior bundle is more vulnerable in distal tears, whereas the posterior band is more frequently affected in proximal tears (41). A popping sensation, or “pop” sign, is more frequent among patients with single tears than among those with attenuated functionally incompetent ligaments without discrete tears (41).
Radiographs depict heterotopic ossification within the UCL in the setting of a chronic tear. Sublime tubercle avulsion fractures may occur and are more likely to require surgical intervention (44). Dynamic valgus stress radiographs depict ulnohumeral gapping, although there is no generally accepted reference value. The ligament is well depicted on US images due to its superficial location. In pitchers, ligament thickening may be adaptive rather than pathologic and can demonstrate hypoechoic foci and/or calcifications (45). Nazarian et al (45) measured the anterior band of the UCL at its midportion, from its superficial surface down to the bone on coronal sections, and reported a mean thickness (± 1 standard deviation) of 6.3 mm ± 1.1 in pitching arms versus 5.3 mm ± 1.0 in nonpitching arms.
Other reported measurements exclude the deep underlying echogenic fat, with mean thicknesses of 1.86 mm (46) and 1.38 mm (47). Valgus stress US can be used to distinguish true laxity from adaptive laxity while directly assessing the ligament. A study of 26 asymptomatic pitchers (45) showed a significant difference in ulnohumeral gapping on valgus stress compared to at rest, with mean differences of 1.4 mm compared to 0.5 mm on the nonthrowing side. A cutoff of 1.5 mm for relative ulnohumeral gapping has been proposed, yielding 81% sensitivity and 91% specificity for UCL injury (Fig 16) (48).
Figure 16. Normal and reconstructed UCLs in two pitchers. ME = medial epicondyle, U = ulna. (A) Longitudinal US image in a 22-year-old asymptomatic right-handed pitcher shows fine homogeneous echogenic striations (arrowheads), indicating an intact ligament; the ulnohumeral gap is also normal, measuring less than 1.5 mm (bracket). (B) Longitudinal US image in another asymptomatic 22-year-old right-handed pitcher after UCL reconstruction shows a markedly thickened and heterogeneous but contiguous reconstructed ligament (arrowheads); there is no gapping with stress (bracket), further supporting an intact ligament.
MRI has high specificity (approaching 100%) but only 57% sensitivity for UCL tears, lower than the sensitivity of CT arthrography (86%) (49). MR arthrography is superior to conventional MRI and CT arthrography, with reported sensitivity and specificity of 100% (50). Similar to other imaging findings in throwers, an abnormal-appearing UCL is not necessarily pathologic. A study of 21 asymptomatic baseball pitchers (28) showed a higher prevalence of increased intrasubstance signal and partial tears in pitching elbows compared to nonpitching elbows.
On coronal MR images, a partial tear at the distal attachment of the anterior bundle appears as a characteristic “T” sign due to fluid or contrast material interposed between the ligament and sublime tubercle (Fig 17). This sign must be differentiated from the normal 3–4-mm separation between the distal insertion of the anterior bundle and the sublime tubercle that is seen in up to 50% of the general population (51). The flexed elbow valgus external rotation (FEVER) view is a more recently described MRI position in which valgus stress is applied to the ulnotrochlear joint (Fig S3) and that yields long-axis images parallel to the UCL (Fig 18) (52). An investigation involving 44 professional baseball pitchers (52) showed high intraobserver agreement on ulnotrochlear joint space measurements and confidence in assessing the UCL with FEVER views, as compared with standard views (Fig 19). The authors found a mean relative increase of 1.80 mm in the ulnotrochlear joint space between the standard and stress FEVER views but found no significant difference between symptomatic and asymptomatic pitchers (52).
Figure 17. Partial tear at the distal attachment of the UCL in a 21-year-old male pitcher with soreness along his medial forearm while throwing. Coronal T1-weighted fat-suppressed MR arthrogram shows interposition of intra-articular contrast material between the ulnar attachment of the UCL and the sublime tubercle, consistent with a partial undersurface tear of the distal anterior bundle of the ligament, or the T sign (arrowheads).
Figure 18. Complete distal UCL tear seen on FEVER MRI view in a 24-year-old professional pitcher with medial elbow pain. (A, B) Standard coronal (A) and FEVER (B) T2-weighted fat-suppressed MR images show a high-grade tear of the distal sublime tubercle attachment of the anterior bundle of the UCL, with focal abnormal signal intensity (arrowhead in A). This is more apparent on the FEVER view, which shows an abrupt change in signal intensity (arrowhead in B). The ulnohumeral gap is normal at rest on the standard coronal view, measuring approximately 3.0 mm between the subchondral bone plates (between arrows on A). On the FEVER view, there is approximately 5.7 mm of gapping between the subchondral bone plates (between arrows on B) of the humeral trochlea and the trochlear groove of the ulna, or a difference of approximately 2.7 mm between the two positions. (C) Intraoperative imaging during surgical exploration and UCL reconstruction confirmed the gapping (straight arrows); the surgically split flexor tendon (curved arrows) is also shown.
Figure 19. Intact postoperative UCL in a 31-year-old left-hand–dominant relief pitcher 13 months after flexor tendon repair and concern for repeat flexor tendon injury. Standard coronal (A) and FEVER (B) T2-weighted fat-suppressed MR images show an intact common flexor tendon with susceptibility artifacts (curved arrow) in the origin, representing a prior flexor tendon repair. There is increased T2 signal intensity in the humeral attachment of the UCL (arrowhead); this is a nonspecific finding that can be seen in intact grafts. Notably, there are no areas of graft discontinuity. The joint width increased from 2.7 mm on the standard view to 4.1 mm on the FEVER view (between straight arrows on B), or a difference of 1.4 mm.
Reconstruction of an injured UCL has become increasingly frequent among elite overhead throwing athletes. UCL injury was once considered to be career ending for professional pitchers until Dr Frank Jobe performed the first successful UCL reconstruction on the pitcher Tommy John in 1974. Between 1974 and 2012, a total of 147 pitchers underwent a single reconstruction procedure, with 80% of them returning to pitch in at least one Major League Baseball game and more than two-thirds of them returning to the same level of competition postoperatively (53). The original fixation described by Jobe and colleagues, involving bone tunnels and a figure-of-8 graft, has evolved into advanced methods, including the docking technique (Fig 20), use of interference screws and suspensory cortical buttons, and hybrid procedures.
Figure 20. UCL reconstruction performed by using the docking technique in a 21-year-old male pitcher with soreness along the medial forearm while throwing (same patient as in Fig 17). Intraoperative photograph showing a flexor-pronator muscle splitting approach (A) and three-dimensional reconstructed CT image (B) show UCL reconstruction performed by using the docking technique. With this method, the UCL is reconstructed by looping a single continuous graft loop (straight arrows, yellow outline in B) through converging ulnar bone tunnels (arrowheads in B), “docking” the sutured graft ends through a single humeral tunnel (curved arrow in B) and tying the two ends of the UCL graft with a suture over a bone bridge (* in B). Note the position of the ulnar tunnels: distal to the joint line and sublime tubercle. The docking technique has become popular, as it minimizes injury to the flexor-pronator mass, avoids the ulnar nerve, and prevents excessive bone resection from the medial epicondyle.
At MRI, an intact graft has low or intermediate signal intensity on T1- and T2-weighted images and may be thickened, especially at its proximal attachment. Less commonly, 25% of intact grafts demonstrate intermediate T1 signal intensity and intermediate to high T2 signal intensity. Graft degeneration has diffuse intermediate signal intensity on T1-weighted images, mild hyperintensity on T2-weighted images, and fibers that cannot be clearly delineated (54). A normal graft may be irregular but should not be wavy (54). In the setting of UCL reconstruction, contrast material extending between the distal reconstructed UCL and the sublime tubercle does not have the same significance (ie, false T sign) because of the tunnel locations approximately 3–4 mm distal to the articular surface (54).
Medial Epicondylitis
Medial epicondylitis (ME), or epicondylosis, colloquially referred to as “golfer’s elbow,” occurs frequently in overhead throwing athletes. The term epicondylitis, which is widely cited in the clinical and imaging literature, is somewhat of a misnomer, as this condition is a noninflammatory process of the tendon rather than the bone and is characterized by tendinosis or degeneration of the flexor-pronator mass. It is due to a repetitive eccentric load of the muscles responsible for wrist flexion and forearm pronation, combined with valgus overload at the elbow during late cocking and early acceleration. The flexor-pronator mass can be hypertrophied in overhead throwing athletes, and it can be accompanied by partial UCL tears, ulnar neuropathy (UN), and lateral-sided impaction injuries (55). Imaging helps to differentiate flexor-pronator mass injuries from other sources of medial elbow pain, given the considerable clinical overlap among these entities.
Radiographs are often normal, although 25%–53% of radiographs may depict calcifications typically at the humeral origin of the common flexor tendon (33%), pronator teres (18%), and UCL attachment (10%) (56). US has demonstrated sensitivity, specificity, and positive predictive and negative predictive values of greater than 90% for ME (57). It shows hypoechoic areas without fiber discontinuity or intratendinous calcifications (tendinosis), a partial tear (focal anechoic area with partial-thickness fiber discontinuity), or a complete tear (full-width tendon gap or tendon nonvisualization) (57). MRI findings include common flexor tendon thickening, increased T1 and T2 signal intensity, peritendinous edema, tendon tear, partial UCL tear, and medial epicondyle marrow edema (58). An MRI case-control study found T2-hyperintense signal of a thickened common flexor tendon and peritendinous edema to be the most specific imaging features (Fig 21) (58).
Figure 21. Medial epicondylitis in a 31-year-old professional baseball pitcher with medial-sided elbow pain. Coronal T2-weighted fat-suppressed MR image shows a high-grade partial-thickness tear of the common flexor tendon at the origin (arrow), with peritendinous edema and chronic adaptive thickening of the UCL (arrowheads).
Ulnar Neuropathy
At the cubital tunnel, the ulnar nerve is bordered by the medial epicondyle anteriorly, cubital retinaculum laterally, and flexor carpi ulnaris posteromedially. UN can occur owing to entrapment, trauma, chronic irritation, and/or compression. During the throwing motion itself, the rapid shift from elbow extension to flexion during acceleration compresses the nerve at the cubital tunnel, with pressures increasing up to 20-fold (40). The nerve can also be irritated by adjacent anterior capsular inflammation, the flexor-pronator muscle mass, injury to the UCL (Fig S4), osteophytes, or avulsed medial epicondyle fragments.
US and MRI are the primary imaging tools for evaluating UN or cubital tunnel syndrome. Commonly described US findings are a swollen hypoechoic nerve and loss of fascicular architecture. There are limited data on the sensitivity and specificity of these findings, although investigators found that combining them with cross-sectional nerve area measurements could yield a sensitivity of 53.7% and a specificity of 95.6% for the diagnosis of cubital tunnel syndrome (59). MRI depicts increased nerve signal intensity on T2-weighted fat-suppressed images, with 83% sensitivity and 85% specificity for diagnosing UN (Fig 22) (60). However, increased nerve signal intensity occurs in 60% of asymptomatic elbows (61) and thus must be correlated with clinical signs or other imaging features of UN.
Figure 22. UN in a 20-year-old male college javelin thrower with a UCL tear and clinical symptoms consistent with ulnar neuritis. Axial (A) and sagittal (B) T2-weighted fat-suppressed MR images show increased signal intensity (arrowheads) within the ulnar nerve at the cubital tunnel and proximal forearm, suggesting neuropathy or neuritis. Compare the signal intensity within the ulnar nerve with the signal intensity of the normal median nerve (arrow in A). The patient underwent subsequent UCL reconstruction and ulnar nerve transposition, with subsequent resolution of the UN symptoms.
Increased cross-sectional nerve area measured 2–3 cm proximal and distal to the cubital tunnel is the most widely accepted imaging feature of the UN cross-sectional area (62,63). Roedl et al (48) proposed different nerve size cutoff values with US and MRI, using 9 mm2 and 10 mm2, respectively, while Terayama et al (63) proposed a nerve size cutoff value of 11 mm2 with both US and MRI. In baseball pitchers, an increase in cross-sectional area was shown to be significantly associated with number of pitches, innings pitched, and games pitched (64). While authors have demonstrated that the cross-sectional area measurements at MRI and US are not significantly different (63), US has the important advantage of dynamic examination. A recent consensus of 15 experts agreed that all investigations of UN should include nerve conduction studies and US assessment of nerve mobility at the elbow, along with cross-sectional area measurements and imaging of the entire nerve (65).
Dynamic examination is performed at the cubital tunnel with passive flexion to approximately 135° from full extension to assess for nerve subluxation or dislocation. During this maneuver, alterations in the bone contours should be avoided, and the nerve should remain posterior to the medial epicondyle. Nerve motion superficial and anterior to the medial epicondyle apex with elbow flexion is often considered abnormal, although this finding can be seen in up to 20% of asymptomatic individuals (66). At dedicated sports medicine facilities and referral centers, use of a combination of modalities is prudent. Concurrent nerve size assessment with US (with a 9-mm2 cutoff value) and either subluxation at US or increased signal intensity at MRI have 90% sensitivity, 100% specificity, and 98% accuracy (48).
Posteromedial Impingement and Valgus Extension Overload Syndrome
Posteromedial impingement and valgus extension overload syndrome are related entities that follow abundant valgus forces from late cocking through deceleration. These valgus stresses result in olecranon wedging into the olecranon fossa, leading to posteromedial impingement.
Figure 23. Graphic illustration of the pathomechanics of valgus extension overload syndrome. This syndrome is characterized by abundant tension on the medial side (black arrows), excessive compression on the lateral side (arrowheads), and shear forces along the head of the olecranon on the posterior side (white arrows), resulting in bone-on-bone contact in the posteromedial aspect of the olecranon (*).
Radiographs and CT images show osteophytes at the posterior olecranon tip and posteromedial olecranon; a fractured osteophyte or intra-articular body may lodge at the olecranon fossa (Fig 24). Chronic mechanical loading due to throwing also results in other adaptive osseous changes. Kooima et al (68) found a significant correlation between subclinical UCL thickening and subchondral sclerosis. In a quantitative CT study, Neil and Schweitzer (69) found significant differences in humeral diaphyseal cortical thickness between throwing versus nonthrowing arms. Apart from showing “kissing” osteochondral lesions of the posteromedial olecranon and trochlea, MRI depicts UCL injuries, olecranon fat pad scarring, medial triceps insertional tendinosis, posteromedial recess synovitis, and olecranon chondral and subchondral abnormalities (28,68,70). MRI findings must be interpreted cautiously, with awareness of the clinical presentation, as 13 of 16 professional baseball players without impingement symptoms had UCL abnormalities and imaging signs of posteromedial impingement in the Kooima et al investigation (68).
Figure 24. Posteromedial impingement with a symptomatic intra-articular body in a 23-year-old professional baseball pitcher with elbow pain mainly with extension. Preoperative sagittal reconstructed (A) and three-dimensional (B) CT images and postoperative three-dimensional CT image (C) show an intra-articular body (arrowhead in A and B) within the olecranon fossa, adjacent sclerosis and cystic changes at the interface of the intra-articular body with the olecranon (curved arrow in A), thickening of the cortical bone in the humeral diaphysis (straight arrows in A), and posteromedial compartment joint space narrowing (between arrows in B and C). CT image after resection of the intra-articular body (C) shows an empty olecranon fossa (arrowhead in C).
Olecranon stress fractures may be seen following posteromedial impingement and are thought to be the result of repetitive triceps contraction during follow-through or from repetitive olecranon impingement. These fractures manifest as transverse or oblique high signal intensity through the affected bone marrow on T2-weighted MR images (Fig 25) (71).
Figure 25. Olecranon stress fracture in a 21-year-old right-handed pitcher, with concern for olecranon stress reaction. (A) Sagittal T2-weighted fat-suppressed MR image shows a faint line with a hyperintense signal (arrow) across the olecranon, with subtle surrounding marrow edema. (B) Sagittal reconstructed CT image confirms the presence of an incomplete stress fracture of the olecranon (arrow). Also note the thickening of the cortical bone in the humeral diaphysis (arrowheads in A and B), which is more readily discernible on the CT image.
Imaging Recommendations in Throwing Athletes
Because most abnormalities in overhead throwing athletes affect the soft-tissue structures of the shoulder, MRI is preferred for comprehensive evaluation after initial radiographic assessment. MR arthrography may be indicated over conventional MRI, given the greater sensitivity of this modality for partial-thickness articular-sided rotator cuff tears and posterosuperior labral disease (72,73). In throwing athletes, MR arthrography may also assist in characterizing GIRD and PSI (15). It should be noted that most of the existing MR arthrography data on throwing injuries were from studies performed on 1.5-T systems, and recommendations will probably evolve with increased accessibility of 3.0-T systems.
In the elbow, MR arthrography also has higher sensitivity and specificity than conventional MRI for evaluation of UCL tears (49,74). Although no comparison between MR arthrography and conventional MRI has been performed in the postoperative setting, Wear et al (54) suggest that MR arthrography delineates the thickening and intermediate graft signal intensity associated with recurrent tears. However, the prediction of valgus laxity is inaccurate with use of both conventional MRI and MR arthrography and requires dynamic stress examination. In addition to using the FEVER MRI view, which increases diagnostic confidence (52), combining MR arthrography with stress US may prove to be useful (49).
Conclusion
Overhead throwing subjects the shoulder and elbow to extreme stresses, predisposing them to various osseous and soft-tissue injuries. Knowledge of the anatomy, biomechanics, and imaging findings of pertinent throwing injuries enables precise diagnosis and timely management.
Presented as an education exhibit at the 2022 RSNA Annual Meeting
References
- 1. . The Incidence of Shoulder Injury among Collegiate Overhead Athletes. J Intercoll Sport 2009;2(2):260–268.
- 2. . Injuries to the shoulder in the throwing athlete. I. Biomechanics/pathophysiology/classification of injury. Am J Sports Med 2000;28(2):265–275.
- 3. . Prevalence and patterns of shoulder injuries in major league baseball. Phys Sportsmed 2020;48(1):63–67.
- 4. . Elbow Injuries in Professional Baseball: Epidemiological Findings From the Major League Baseball Injury Surveillance System. Am J Sports Med 2017;45(10):2319–2328.
- 5. . The biomechanics of throwing: simplified and cogent. Sports Med Arthrosc Rev 2014;22(2):72–79.
- 6. . Biomechanics of the elbow in the throwing athlete. Oper Tech Sports Med 1996;4(2):62–68.
- 7. . Shoulder injuries in the overhead-throwing athlete: epidemiology, mechanisms of injury, and imaging findings. Radiology 2018;286(2):370–387.
- 8. . Elbow injuries in the throwing athlete. Orthop Clin North Am 2014;45(4):571–585.
- 9. . The disabled throwing shoulder: spectrum of pathology. I. Pathoanatomy and biomechanics. Arthroscopy 2003;19(4):404–420.
- 10. . Biomechanics of the elbow during baseball pitching. J Orthop Sports Phys Ther 1993;17(6):274–278.
- 11. . Glenohumeral internal rotation deficit: pathogenesis and response to acute throwing. Sports Med Arthrosc Rev 2012;20(1):34–38.
- 12. . Glenohumeral range of motion deficits and posterior shoulder tightness in throwers with pathologic internal impingement. Am J Sports Med 2006;34(3):385–391.
- 13. . Stretching the Posterior Shoulder Structures in Subjects with Internal Rotation Deficit: Comparison of Two Stretching Techniques. Shoulder Elbow 2012;4(1):56–63.
- 14. . Posterior capsular fibrosis in professional baseball pitchers: case series of MR arthrographic findings in six patients with glenohumeral internal rotational deficit. Clin Imaging 2007;31(5):343–348.
- 15. . Shoulder MR arthrography of the posterior labrocapsular complex in overhead throwers with pathologic internal impingement and internal rotation deficit. Skeletal Radiol 2007;36(6):495–502.
- 16. . Arthroscopic removal of symptomatic Bennett lesions in the shoulders of baseball players: arthroscopic Bennett-plasty. Am J Sports Med 2002;30(5):728–736.
- 17. . Infraspinatus atrophy due to Bennett lesion causing suprascapular nerve palsy. J Ultrason 2021;21(85):e177–e181.
- 18. . Infraspinatus muscle atrophy in professional baseball players. Am J Sports Med 2004;32(1):116–120.
- 19. . Bennett lesions in baseball players detected by magnetic resonance imaging: assessment of association factors. J Shoulder Elbow Surg 2016;25(5):730–738.
- 20. . Prevalence of the Bennett lesion of the shoulder in major league pitchers. Am J Sports Med 2004;32(1):121–124.
- 21. . Posterior shoulder pain in throwing athletes with a Bennett lesion: factors that influence throwing pain. J Shoulder Elbow Surg 2006;15(1):72–77.
- 22. . Impingement of the deep surface of the supraspinatus tendon on the posterosuperior glenoid rim: an arthroscopic study. J Shoulder Elbow Surg 1992;1(5):238–245.
- 23. . Microinstability and internal impingement of the shoulder. Semin Musculoskelet Radiol 2015;19(3):277–283.
- 24. . Posterosuperior glenoid impingement of the shoulder: findings at MR imaging and MR arthrography with arthroscopic correlation. Radiology 1994;193(2):431–436.
- 25. . Internal impingement of the shoulder: comparison of findings between the throwing and nonthrowing shoulders of college baseball players. Arthroscopy 1999;15(3):253–258.
- 26. . Mimickers of Hill-Sachs Lesions. Can Assoc Radiol J 2021;72(2):258–270.
- 27. . Acute shoulder trauma: what the surgeon wants to know. RadioGraphics 2015;35(2):475–492.
- 28. . Three-tesla MR imaging of the elbow in non-symptomatic professional baseball pitchers. Skeletal Radiol 2015;44(1):115–123.
- 29. . Magnetic resonance imaging of the asymptomatic shoulder of overhead athletes: a 5-year follow-up study. Am J Sports Med 2003;31(5):724–727.
- 30. . The disabled throwing shoulder: spectrum of pathology. III. The SICK scapula, scapular dyskinesis, the kinetic chain, and rehabilitation. Arthroscopy 2003;19(6):641–661.
- 31. . Effect of posterior shoulder tightness on internal impingement in a cadaveric model of throwing. Knee Surg Sports Traumatol Arthrosc 2015;23(2):548–554.
- 32. . The peel-back mechanism: its role in producing and extending posterior type II SLAP lesions and its effect on SLAP repair rehabilitation. Arthroscopy 1998;14(6):637–640.
- 33. . Type II SLAP lesions: three subtypes and their relationships to superior instability and rotator cuff tears. Arthroscopy 1998;14(6):553–565.
- 34. . Magnetic resonance appearance of posterosuperior labral peel back during humeral abduction and external rotation. Skeletal Radiol 2010;39(1):19–26.
- 35. . Throwing-related injuries of the subscapularis in professional baseball players. Skeletal Radiol 2016;45(1):41–47.
- 36. . Treating Subscapularis and Lesser Tuberosity Avulsion Injuries in Skeletally Immature Patients: A Systematic Review. Arthroscopy 2016;32(5):919–928.
- 37. . Latissimus dorsi and teres major tears in professional baseball pitchers: a case series. Am J Sports Med 2009;37(10):2016–2020.
- 38. . Surgical anatomy of latissimus dorsi muscle in transfers about the shoulder. Am J Orthop 2009;38(3):E64–E67.
- 39. . Latissimus dorsi tendinosis and tear: imaging features of a pseudotumor of the upper limb in five patients. AJR Am J Roentgenol 2005;185(5):1145–1151.
- 40. . Pathologies of the shoulder and elbow affecting the overhead throwing athlete. Skeletal Radiol 2017;46(7):873–888.
- 41. . The pathoanatomy of the anterior bundle of the medial ulnar collateral ligament. J Shoulder Elbow Surg 2019;28(8):1497–1504.
- 42. . Elbow valgus stability of the transverse bundle of the ulnar collateral ligament. BMC Musculoskelet Disord 2021;22(1):873.
- 43. . Stress ultrasound evaluation of medial elbow instability in a cadaveric model. Am J Sports Med 2014;42(10):2463–2469.
- 44. . Avulsion fracture of the ulnar sublime tubercle in overhead throwing athletes. Am J Sports Med 2002;30(3):426–431.
- 45. . Dynamic US of the anterior band of the ulnar collateral ligament of the elbow in asymptomatic major league baseball pitchers. Radiology 2003;227(1):149–154.
- 46. . US of the anterior bundle of the ulnar collateral ligament: findings in five cadaver elbows with MR arthrographic and anatomic comparison: initial observations. Radiology 2003;227(2):561–566.
- 47. . Ultrasound evaluation of the ulnar collateral ligament of the elbow: which method is most reproducible? Skeletal Radiol 2017;46(8):1081–1085.
- 48. . Potential Utility of a Combined Approach with US and MR Arthrography to Image Medial Elbow Pain in Baseball Players. Radiology 2016;280(1):328.
- 49. . Preoperative evaluation of the ulnar collateral ligament by magnetic resonance imaging and computed tomography arthrography: evaluation in 25 baseball players with surgical confirmation. Am J Sports Med 1994;22(1):26–32.
- 50. . Accuracy of 3-T MR arthrography versus conventional 3-T MRI of elbow tendons and ligaments compared with surgery. AJR Am J Roentgenol 2015;204(1):W70–W75.
- 51. . Anterior bundle of ulnar collateral ligament: evaluation of anatomic relationships by using MR imaging, MR arthrography, and gross anatomic and histologic analysis. Radiology 2004;231(3):797–803.
- 52. . FEVER: The Flexed Elbow Valgus External Rotation View for MRI Evaluation of the Ulnar Collateral Ligament in Throwing Athletes—A Pilot Study in Major League Baseball Pitchers. AJR Am J Roentgenol 2021;217(5):1176–1183.
- 53. . Performance, return to competition, and reinjury after Tommy John surgery in major league baseball pitchers: a review of 147 cases. Am J Sports Med 2014;42(6):1323–1332.
- 54. . MRI of the reconstructed ulnar collateral ligament. AJR Am J Roentgenol 2011;197(5):1198–1204.
- 55. . In vivo three-dimensional mechanical actions of individual. J Appl Biomech 2008;24(4):325–332.
- 56. . Radiologic evaluation and clinical effect of calcification in medial epicondylitis. J Shoulder Elbow Surg 2022;31(2):375–381.
- 57. . Diagnostic value of ultrasonography for clinical medial epicondylitis. Arch Phys Med Rehabil 2008;89(4):738–742.
- 58. . Magnetic resonance imaging findings in patients with medial epicondylitis. Skeletal Radiol 2005;34(4):196–202.
- 59. . The validity of ultrasonographic assessment in cubital tunnel syndrome: the value of a cubital-to-humeral nerve area ratio (CHR) combined with morphologic features. Ultrasound Med Biol 2010;36(3):376–382.
- 60. . Ulnar neuropathy at the elbow: MR neurography—nerve T2 signal increase and caliber. Radiology 2011;260(1):199–206.
- 61. . Ligaments and plicae of the elbow: normal MR imaging variability in 60 asymptomatic subjects. Radiology 2010;257(1):185–194.
- 62. . Ultrasound in the diagnosis of ulnar neuropathy at the cubital tunnel. J Hand Surg Am 2006;31(7):1088–1093.
- 63. . Optimal Measurement Level and Ulnar Nerve Cross-Sectional Area Cutoff Threshold for Identifying Ulnar Neuropathy at the Elbow by MRI and Ultrasonography. J Hand Surg Am 2018;43(6):529–536.
- 64. . Pre- and Postseason Dynamic Ultrasound Evaluation of the Pitching Elbow. Arthroscopy 2015;31(9):1708–1715.
- 65. . Expert consensus on the combined investigation of ulnar neuropathy at the elbow using electrodiagnostic tests and nerve ultrasound. Clin Neurophysiol 2021;132(9):2274–2281.
- 66. . Sonography of common peripheral nerve disorders with clinical correlation. J Ultrasound Med 2016;35(4):683–693.
- 67. . Elbow medial ulnar collateral ligament insufficiency alters posteromedial olecranon contact. Am J Sports Med 2004;32(7):1607–1612.
- 68. . Evidence of subclinical medial collateral ligament injury and posteromedial impingement in professional baseball players. Am J Sports Med 2004;32(7):1602–1606.
- 69. . Humeral cortical and trabecular changes in the throwing athlete: a quantitative computed tomography study of male college baseball players. J Comput Assist Tomogr 2008;32(3):492–496.
- 70. . Posteromedial elbow impingement: magnetic resonance imaging findings in overhead throwing athletes and results of arthroscopic treatment. Arthroscopy 2011;27(10):1364–1370.
- 71. . Correlation of MRI to Arthroscopy in the Elbow: Thrower’s Elbow and Ulnar Collateral Ligament Injury. Sports Med Arthrosc Rev 2017;25(4):191–198.
- 72. . Shoulder MR arthrography: which patient group benefits most? AJR Am J Roentgenol 2004;183(4):969–974.
- 73. . Society of skeletal radiology position paper: recommendations for contrast use in musculoskeletal MRI—when is non-contrast imaging enough? Skeletal Radiol doi: https://doi.org/10.1007/s00256-023-04367-8. Published online June 10,2023.
- 74. . Diagnostic Imaging of Ulnar Collateral Ligament Injury: A Systematic Review. Am J Sports Med 2020;48(11):2819–2827.
Article History
Received: Apr 30 2023Revision requested: May 20 2023
Revision received: June 22 2023
Accepted: June 28 2023
Published online: Nov 02 2023
