- The shoulder is one of the most commonly imaged joints in the body, surpassed only by the knee. Imaging plays an important role in the diagnosis of the various types of shoulder pathology, providing useful information that helps guide appropriate treatment selection. The 1st imaging method employed is typically radiographs, followed by MR, ultrasound, &/or CT, depending on the patient's history and physical examination findings. This section will review the imaging appearances of various types of soft tissue and bone pathologies seen in the setting of shoulder trauma, including injuries to the rotator cuff, biceps tendon, glenoid labrum, humerus, and scapula.
Terminology and Conventions
- Rotator cuff tendon pathology is one of the most common reasons for shoulder imaging, especially in older patients. The term tendinopathy is preferred to tendinosis or tendinitis to describe the chronic breakdown/remodeling that occurs over time secondary to microtrauma and degeneration. Similarly, calcific tendinopathy is a more correct term for the changes found in the setting of calcium hydroxyapatite deposition than terms, such as calcific tendinitis, which is typically seen in the surgical literature.
- The terminology describing rotator cuff tendon tears is somewhat unique and typically not used (or expected) for the description of tendon tears in other parts of the body. Supraspinatus, infraspinatus, subscapularis, and teres minor tendon injuries are most commonly described in terms of partial-thickness or full-thickness tearing depending on the presence and extent of tearing involving the tendon from its articular surface to its bursal surface. When the entire tendon attachment is torn, then the term full width or complete may be added to the tendon description. If 2 tendons are torn in this manner, then the tearing can be described as a massive tear, most commonly involving the supraspinatus and infraspinatus tendons.
- Glenoid capsulolabral pathology is another common reason for shoulder imaging, especially in younger patients with shoulder instability. Different terminology, such as Bankart tears, anterior labroligamentous periosteal sleeve avulsion (ALPSA) lesions, superior labral anterior to posterior (SLAP) tears, and glenoid labrum articular disruption (GLAD) lesions, have been used to describe the various types of capsulolabral pathology typically seen. The use of these acronyms to describe labroligamentous injuries can be challenging but useful in terms of understanding the underlying pathologies and imaging findings.
- Labral tear classification schemes typically are based on location because the location often correlates best with patient symptoms and predicts prognosis. There are 2 main classifications. First, labral tears can be described based on quadrants: anterosuperior, anteroinferior, posterosuperior, and posteroinferior. Second, one could use a clockface scheme in which 12 o'clock refers to the position of the biceps anchor (superior), and 6 o'clock corresponds to the inferior margin of the glenoid at the long head triceps tendon origin. The location of 3 o'clock can vary depending on the surgeon, but it is described most commonly along the anterior glenoid. Some authors reverse the 3 o'clock convention, however, depending on whether the right or left shoulder is involved. Similarly, the 9 o'clock position most commonly denotes the posterior labrum, but this also could vary.
- Although MR and CT imaging sequences usually are prescribed in oblique planes reflecting the angle of the scapula relative to the body, it is conventional to refer to these oblique imaging planes as coronal and sagittal in discussion of the shoulder.
- The glenohumeral articulation provides the greatest ROM of any joint in the body. It is a ball-and-socket joint with a shallow socket (glenoid fossa) articulating with a curved surface (humeral head), similar to a golf ball on a tee. The coracoacromial arch provides an osseous restraint to humerus motion superiorly. Glenohumeral stability is maintained by the coordinated mechanism of soft tissue and osseous structures, including
- The glenoid labrum is a triangular structure in cross section that extends around the periphery of the osseous glenoid rim. The labrum is made of fibrous tissue except for a small fibrocartilage transition zone at its attachment to the hyaline cartilage of the glenoid. Because the labrum is fibrous, its shape can be variable on MR (e.g., triangular or rounded). Although the labrum acts as a bumper to deepen the glenoid socket, its main role is as a continuation of the joint capsule and the insertion site of the glenohumeral ligaments.
- The rotator cuff muscles consist of the supraspinatus, infraspinatus, teres minor, and subscapularis. The supraspinatus tendon inserts on the superior facet and anterior portion of the middle facet of the humeral greater tuberosity. The infraspinatus tendon merges with the posterior supraspinatus to insert along the entire middle facet of the greater tuberosity. The teres minor tendon inserts along the inferior facet. The subscapularis tendon inserts along the lesser tuberosity with a smaller muscular insertion found at the humeral neck. These tendons form a cuff of continuous tissue in their distal 1.5 cm that acts to abduct and internally and externally rotate the humeral head while stabilizing the glenohumeral joint.
- The long head biceps tendon crosses the glenohumeral joint and acts as a dynamic stabilizer and forward flexor of the joint. The extraarticular portion of the tendon ascends in the bicipital groove between the lesser tuberosity and the anterior aspect of the greater tuberosity before exiting the groove and curving medially within the biceps pulley sling. When the tendon exits the pulley, it is oriented horizontally and becomes intraarticular. The tendon has a variable origin from the superior labrum and supraglenoid tubercle, though it also may originate directly from the joint capsule.
- The intraarticular portion of the humeral head is predominantly covered by hyaline cartilage, except for the bare area posteriorly. The glenoid fossa has a central bare spot that should not be confused with a pathologic focal cartilage defect.
- Shoulder injuries can be secondary to a variety of etiologies, including acute trauma, overuse, and age-related injuries. Fractures around the shoulder are common after trauma. Posttraumatic joint instability frequently occurs at the glenohumeral joint and at the acromioclavicular joint. Posttraumatic anterior shoulder instability is most common at the glenohumeral joint. Acromioclavicular joint instability injuries are typically referred to as separations. Overuse injuries (e.g., internal impingement) are common in overhead-throwing athletes and can result in decreased throwing velocity. Rotator cuff tendon injuries usually are a result of chronic breakdown and degeneration, setting the stage for tendon tearing, but rarely also can occur in the setting of acute trauma.
- Several commonly encountered normal variants occur around the shoulder, including the os acromiale and sublabral foramen, that are usually asymptomatic. Conversely, rotator cuff tears may be clinically asymptomatic. The tendon has few nerve fibers, and most of the pain associated with cuff pathology likely arises from synovitis or mechanical catching of redundant synovium by the tear.
- The imaging algorithm for the vast majority of shoulder pathologies typically begins with radiographs. The standard series often includes an anteroposterior (AP) internal rotation view, Neer AP (or Grashey) view with external rotation, and an axillary view. Some prefer an AP external rotation view, whereas others obtain a Zanca AP view, which has 10-20° of cephalad tilt to better visualize the acromioclavicular joint. A scapular-Y view may be obtained for certain fractures or in patients who cannot abduct the arm for an axillary view. The arch view is useful for determining the shape of the anterior acromion. The West Point view profiles the anteroinferior glenoid rim, increasing the sensitivity for detection of anterior glenoid bone loss &/or fractures. Multiple additional specialty views, including the Stryker notch and Rockwood views, may be helpful for profiling specific shoulder pathology.
- CT scans are helpful for characterizing complex scapula and proximal humerus fractures, especially for assessing involvement of the articular surfaces. Ultrasound is increasingly used in the shoulder to evaluate the rotator cuff and biceps tendons and to guide therapeutic injections. Most practices no longer perform shoulder arthrography except as a precursor to MR or CT arthrography. Although full-thickness rotator cuff tears can be diagnosed with conventional arthrography, tear size and morphology are better evaluated with an accompanying MR or CT exam.
- MR is performed with the patient supine and the humerus comfortably externally rotated. The main challenges associated with shoulder MR are motion artifact and poor signal-to-noise ratio. The anatomy of the shoulder prevents shoulder coils from either rigidly immobilizing the joint (to prevent motion artifact from breathing) or encircling the joint (to provide uniform signal intensity across the imaging volume). Image quality is generally better on 3T scanners and with more rigid 8-channel or higher shoulder coils.
- No single preferred MR protocol exists for imaging the shoulder, but most radiologists agree with several basic principles. Because the scapular blade is angled 30° anteriorly relative to the body, the standard coronal and sagittal imaging planes are oriented relative to the glenoid articular surface or supraspinatus tendon. The oblique coronal plane is typically oriented perpendicular to the glenoid fossa or parallel to the long axis of the supraspinatus tendon. The oblique sagittal plane is oriented parallel to the glenoid face (or orthogonal to the oblique coronal plane). Axial images are usually acquired without anatomic angulation.
- Two other slice positioning techniques can sometimes be helpful and deserve mention. Because the scapula is also tilted anteriorly in the sagittal plane, angling the axial images perpendicular to the true vertical axis of the glenoid fossa better profiles the 3 o'clock and 9 o'clock positions of the labrum and reduces the partial volume averaging of anteroinferior and posterosuperior labral tears. Second, adding angled oblique sagittal images perpendicular to the distal-most fibers of the supraspinatus tendon can better demonstrate tendon tears that commonly occur in this region.
- Because labral tears at times can be difficult to see with conventional MR techniques, direct MR arthrography is used in the setting of glenohumeral instability. MR arthrography has higher accuracy for delineation of labroligamentous pathology compared to routine shoulder MR imaging due to joint distension and increased contrast:noise afforded by the injectate. Abduction external rotation (ABER) images can further improve the accuracy of MR arthrography for certain pathology. The ABER position pulls the anterior band of the inferior glenohumeral ligament taut, which can distract, and, therefore, make more apparent, focal Bankart tears and Bankart variations.
- Multiple MR pulse sequences have been advocated for imaging of the shoulder. High accuracy for cuff tears has been reported with fat-suppressed fast spin-echo T2-weighted sequences in the oblique coronal and sagittal planes, often with a TE of 45-55 msec at 1.5T. It is also helpful to obtain at least 1 T1- or proton density-weighted sequence without fat suppression to evaluate for fatty muscle atrophy. The sequences used for axial images are more variable with some preferring a fat-suppressed intermediate-weighted sequence (TE = 30-40 msec at 1.5T), while others obtain T2*-weighted gradient-echo &/or proton density-weighted sequences. The choice of sequences generally depends on personal preference and machine capabilities. Additional sequences, such as 3D-Dixon sequences, are being used to create 3D MR models to evaluate the glenoid and humeral head in the setting of shoulder instability.
- Goes PCK et al: Radiographic/MR imaging correlation of the shoulder. Magn Reson Imaging Clin N Am. 27(4):575-85, 2019
- Roy JS et al: Diagnostic accuracy of ultrasonography, MRI and MR arthrography in the characterisation of rotator cuff disorders: a meta-analysis. Br J Sports Med. 49(20):1316-28, 2015
- Nazarian LN et al: Imaging algorithms for evaluating suspected rotator cuff disease: Society of Radiologists in Ultrasound consensus conference statement. Radiology. 267(2):589-95, 2013
- Smith TO et al: A meta-analysis of the diagnostic test accuracy of MRA and MRI for the detection of glenoid labral injury. Arch Orthop Trauma Surg. 132(7):905-19, 2012
- Cowderoy GA et al: Overuse and impingement syndromes of the shoulder in the athlete. Magn Reson Imaging Clin N Am. 17(4):577-93, v, 2009
- Murray PJ et al: Clinical update: MR imaging of the shoulder. Sports Med Arthrosc. 17(1):40-8, 2009
- Steinbach LS: MRI of shoulder instability. Eur J Radiol. 68(1):57-71, 2008
Related Differential Diagnoses