The shoulder girdle comprises three bones (the proximal humerus, the scapula, and the clavicle) that articulate in three joints: the glenohumeral, acromioclavicular, and sternoclavicular joints. The head of the humerus is much larger than the glenoid fossa, giving the glenohumeral joint a wide range of movement at the cost of instability. Static (glenoid labrum, capsule, glenohumeral and coracoacromial ligaments) and dynamic stabilizers (rotator cuff tendons) maintain the joint congruence during movement(1–3). The rotator cuff is composed of four muscles with relative tendons attaching onto the humerus: the subscapularis, the supraspinatus, the infraspinatus, and the teres minor. The long head of the biceps tendon (LHBT) has a proximal insertion at the apex of the glenoid (bicipital-labral complex), courses laterally and anteriorly through the so-called rotator interval (between the subscapularis and supraspinatus tendons), and turns down vertically through the bicipital groove of the humerus (Fig. 1). The LHBT is the only tendon around the shoulder with a synovial sheath, which communicates with the glenohumeral joint space.
The subacromial-subdeltoid (SASD) bursa is a large synovial space between the coracoacromial arch and the supraspinatus tendon (1–2 mm in normal thickness) which facilitates motion and dissipates the friction caused by complex shoulder movements(4).
Due to their superficial location, soft tissue structures of the shoulder can be easily scanned. Nevertheless, the expertise of the operator performing the ultrasonography (US) examination is a relevant aspect which may significantly impact diagnostic accuracy. In fact, several pitfalls related to the ultrasound technique may mislead inexperienced US operators, such as anisotropy artefacts, causing an artifactually hypoechoic tendon appearance that may simulate pathology. Therefore, a scanning protocol is highly recommended in order to reduce the rate of operator errors by following a standardized scheme that includes a list of primary structures. This method is crucial for an exhaustive and efficient examination, also because focal shoulder symptoms do not always correlate with the location of the disease(2).
Several technical guidelines have been issued throughout the years by different scientific societies. The three main guidelines are: EULAR (European League Against Rheumatism), ESSR (European Society of Musculoskeletal Radiology), and the American College of Radiology (ACR)/American Institute of Ultrasound in Medicine (AIUM)/Society of Pediatric Radiology (SPR)/Society of Radiologists in Ultrasound (SRU)(5). They propose relatively uniform approaches, with some differences related to the position of the examiner (standing in front or behind the patient) and patient positioning (Crass or modified Crass) for supraspinatus tendon assessment(2).
The shoulder US examination is typically performed using high-frequency (6–15-MHz) linear broadband array transducers. The patient is preferably imaged while seated, with the radiologist either seated in front of or standing behind the patient. In general, each anatomic structure is evaluated in orthogonal planes by asking the patient to perform specific positional maneuvers, as needed. An exhaustive examination should include evaluating the LHBT, rotator cuff tendons, acromioclavicular joint and SASD bursa. Moreover, dynamic scans may allow the assessment of possible shoulder impingement. Detailing shoulder scanning technique is beyond the purpose of this article, but in our practice we usually refer to the ESSR Technical Guidelines(6).
All shoulder tendons are examined in short and long axis, sometimes with gentle toggling of the transducer to eliminate the anisotropy artefact that may mimic tendinopathy or a tear. The LHBT is usually scanned from the proximal (superior aspect of the bicipital groove) to the distal (pectoralis major tendon-humerus attachment) aspect, firstly in the transverse plane. Dynamic US of the LHBT is performed to evaluate for possible tendon subluxation or dislocation(7). The main clinical indications for LHBT include tendinosis/tenosynovitis, rupture, and subluxation or dislocation.
LHBT
They commonly occur at the entrance to the humeral groove and may propagate distally or proximally, where they may extend into the biceps anchor with associated superior labral tears, also known as the SLAP lesion (superior labral tear from anterior to posterior). Of note, although MR arthrography remains the technique of choice for evaluating SLAP lesions, US also showed some potential for detecting SLAP lesions in a pivotal study by Alali
In the setting of
Rotator cuff (RC) is composed of four fibrous tendons (subscapular, supraspinatus, infraspinatus, teres minor), which appear on US evaluation as hyperechoic structures with a convex surface and uniform fibrillar appearance(10). RF pathologies include a broad spectrum of diseases, including tendinopathy, tendon tears (full- and partial-thickness tears), calcific depositions or calcific enthesopathy(7).
The appearance of RF
With expert operators, it has been reported to be as accurate as magnetic resonance imaging (MRI)(5,11,12). Partial thickness tears may involve the articular or bursal surface, and are associated with cortical irregularity (“pitting”) at the tendon insertion (Fig. 3).
RC tears should be described in terms of their location and dimensions in short axis (e.g. anterior, middle or posterior fibers) and long axis (e.g. involving the footprint or musculotendinous junction), their shape, and tendon retraction(7). A massive tear is defined as greater than 5 cm in width and/or involving two or more tendons. Chronic full-thickness tears are commonly associated with tendon retraction and less commonly with joint or bursal effusion, which is typically seen in acute injury (Fig. 4). Other signs suggesting an RC tear include bony cortical irregularities of the footprint, the “cartilage interface sign”, glenohumeral joint and SASD bursa effusion, as well as various degrees of SASD bursa wall depression in the location of the tear. Cortical irregularity and joint effusion are the signs with the highest values of sensitivity, specificity, and positive and negative predictive values for US detection of supraspinatus tendon tears(13). The cartilage interface sign is a curvilinear hyperechoic line that courses parallel to the hypoechoic hyaline cartilage of the humeral head, and is located at the interface between the hyaline cartilage and the abnormal hypoechoic tendon (Fig. 5). It results from an increased US transmission due to changes in acoustic impedance in cases of articular surface–sided tendon disease, and it is most pronounced in full-thickness RC tears(7). Complete evaluation of RC tears also includes an assessment of muscle trophism. In fact, fatty degeneration in the setting of a tendon tear is a negative prognostic factor in the subsequent tendon repair. US and MR imaging have comparable diagnostic performance in the detection of rotator cuff atrophy; on US, fatty degeneration appears as increased echogenicity and reduced muscle volume(14).
Patients of 30–50 years of age are the most prone. Although the exact pathogenesis is still debated, it is probably a multifactorial disease which occurs in multiple stages: the pre-calcific, calcific (including formative and resorptive phases) and post-calcific. Pain is associated with the resorptive phase(16). US is useful for the detection and localization of calcifications within the tendon, which appear as fluffy or well-defined hyperechoic deposits, with posterior acoustic shadowing in cases of hard calcifications (Fig. 6). Color Doppler US is helpful in detecting the resorptive phase, which is associated with an increased Doppler signal. Besides diagnostic purposes, US is also capable of guiding therapeutic needle placement and irrigation for symptomatic calcific tendinitis(17).
Conditions that can affect the SASD bursa include bursal effusion secondary to rotator cuff disease, infection, and inflammatory bursitis.
The acromioclavicular (AC) joint is a diarthrodial synovial joint. The articular surfaces, encased in a fibrous capsule, are separated by a fibrocartilaginous disk(2). Common clinical indications for the evaluation of the AC joint include osteoarthrosis, acute trauma (i.e. separation or dislocation), synovitis, synovial cysts, osteolysis, and needle guidance for aspiration or injection(7).
The primary role of US is to evaluate capsular hypertrophy and distension. A capsule-to-bone distance less than 3 mm rules out synovial hypertrophy and joint effusion (Fig. 7)(20).
In contrast, a 2–3 mm comparative difference between AC joints width – bilaterally evaluated – is considered abnormal in the appropriate clinical setting and when symptomatic(21). In addition, even though US is not primarily used in the setting of trauma, it is more sensitive than radiography for the identification of grade I AC joint injury (soft tissue swelling and capsular distension at US) and has the same accuracy as radiography in more severe injuries (Fig. 8)(7).
Impingement is a clinical scenario involving painful functional limitation of the shoulder, thought to be secondary to the compression or altered dynamics that irritate and ultimately damage the tissue around the shoulder joint(22). External impingement, which relates to abnormal contact between the humeral head and extra-articular structures such as the acromion (subacromial impingement) and the coracoid process (subcoracoid impingement), is better evaluated with dynamic maneuvers(22).
Indications include cysts of the spinoglenoid notch (SGN), glenohumeral joint degeneration, and glenohumeral joint synovitis. The SGN is formed lateral to where the spine extrudes from the scapula and corresponds to the site where the suprascapular nerve passes around the scapula. The nerve can be occasionally compressed at SGN by ganglia arising from the glenohumeral joint (e.g. due to glenoid-labral tears), possibly leading to infraspinatus and supraspinatus progressive fatty degeneration (Fig. 10).
Although US is not the primary modality for evaluating glenohumeral joint degenerative disease, US findings in cases of severe degeneration include severe joint space narrowing, bulky osteophytosis, cortical irregularity, and eventually degenerative tearing of the posterior labrum and echogenic joint bodies or debris(7). Glenohumeral joint synovitis is seen on US as anechoic joint effusion with synovial thickening involving the posterior and axillary joint recesses, with an associated increase in US signal on Color Doppler(7). The location of the axillary and suprascapular nerves can be seen with US in limited regions of the shoulder. The more important role of US is detecting indirect signs of nerve pathology, such as muscle atrophy (especially without signs of tendon tear), and possible sources of neuropathy, such as compression from a mass or cyst(24).
Elastography is based on the principle of mechanical stress causing changes in stiffness to the underlying tissue. Two techniques are used in clinical practice: strain and shear-wave elastography. The latter is the less operator-dependent of the two, allowing a quantitative evaluation based on focused acoustic radiation force, which provides local stress and generates tissue displacement represented a color map(25,26).
Proposed clinical applications of elastography include the detection and quantification of fatty muscle degeneration (especially after a tendon tear), assessment of tendon stiffness, examination of trapezius and deltoid muscles (important for pre- and post-operative evaluations), and assessment of posterior capsule and coracohumeral ligament stiffness(26). However, there is no standardized guided technique, and in the majority of cases the examination is highly operator-dependent. A minimum distance (usually 1.2 mm) between the skin and examined tissue is needed to place the elastogram box. This may require the application of gel pads or adapters in slim individuals or for the evaluation of very superficial structures. In addition, the exact orientation of the transducer is essential for generating reproducible results. Finally, muscle stiffness is also strongly affected by the grade of muscle contraction(26).
According to the current literature, only limited data are available so far, based on non-controlled studies with small case numbers. Therefore, further investigations are needed.
The use of CEUS has been proposed for with the evaluation of adhesive capsulitis(27). Adhesive capsulitis, also referred to as frozen shoulder, presents with an insidious onset and shoulder pain with a range of motion globally restricted. Accurate diagnosis can be challenging because imaging findings are usually unremarkable(27). According to the literature, there have been few studies investigating AC with CEUS after the administration of microbubble-based ultrasound contrast agents (SonoVue), both intravenously (to facilitate microcirculation detection) and intraarticularly (US arthrography)(28). Filling defects and enhanced synovial microcirculation of the joint cavity may be considered a useful sign to indicate AC. However, only limited data are available, and further investigations are needed. Another study reported the use of CEUS for preoperative deltoid assessments, as a predictor of shoulder dysfunction after reverse shoulder arthroplasty (RSA)(29). The deltoid, in fact, represents the main muscle for elevation and abduction after RSA. CEUS allows dynamic quantification of perfusion in muscle and, therefore, represents a functional real-time biomarker of muscle vitality. Perfusion of the deltoid quantified by CEUS significantly correlated with postoperative shoulder function. Preoperative deltoid dynamic perfusion (PE) revealed a significant correlation with deltoid function after RSA. It might also be useful to detect adaptation processes of the deltoid after RSA, without the drawback of MR metal artifacts(29).
Among techniques for characterizing contractile properties of muscle tissue, US technique equipped with automatic speckle-tracking software is a non-invasive method for determining the contractile properties of muscle tissue. It has been used for measuring muscle strain
US is an extremely valuable imaging technique in daily clinical practice, and it is considered as the first-choice technique to assess tendons, burse and capsuloligamentous structures. Moreover, the advent of ultra-high-frequency US (UHFU), with probes up to 70 MHz allowing for a resolution as low as 30 μm,(31) represents a promising possibility for an improved evaluation of the shoulder anatomy and diagnostic and therapeutic strategies.