Abstract
The rotator cuff is a group of four muscles and tendons surrounding the shoulder joint providing it strength and stability. The rotator cuff consists of the subscapularis, supraspinatus, infraspinatus and teres minor. Many shoulder complaints are caused by rotator cuff pathology such as impingement syndrome, tendon tears and other diseases e.g. calcific tendonitis. Diagnosis starts with clinical history and physical examination, after which imaging is often used to help confirm clinical findings depending on the differential diagnosis. The aim of the article is to review the frequently used imaging modalities to assess the rotator cuff and cuff-related disease, specifically focusing on radiography, ultrasonography and magnetic resonance imaging. This article will outline the advantages and disadvantages for each modality and illustrate typical radiological findings of common rotator cuff pathologies.
Keywords: Rotator cuff tendons, Imaging, Shoulder pain, MRI, Ultrasound
1. Introduction
The rotator cuff (RC) is a group of four muscles and tendons surrounding the shoulder joint providing it strength and stability. The RC consists of subscapularis (SSC), supraspinatus (SST), infraspinatus (IST) and teres minor (TM). Many shoulder complaints are caused by RC pathology such as impingement syndrome, tears and calcific tendonitis. Diagnosis starts with clinical history and physical examination, after which radiography is typically the initial imaging deployed due to its accessibility and low cost. Radiographs provide indirect information on the condition of the RC and shoulder joint which can then be correlated with ultrasound (US) and/or magnetic resonance (MR) imaging.1 In this paper, we will discuss the role of these three main modalities in the assessment of the rotator cuff and its contiguous structures.
2. Radiography
The number of radiographic views performed depends on department protocols and the clinical indications for imaging. Standard projections obtained will vary between different local protocol but most institutions would typically perform two orthogonal projections of the shoulder. Commonly obtained projections include the anteroposterior (AP) and axillary view. The AP view can also be obtained in internal or external rotation depending on the clinical indication. Each projection has its advantages in evaluation of different pathologies of the shoulder (Fig. 1, Fig. 2, Fig. 3, Fig. 4). To obtain the most information from these radiographs, meticulous radiographic technique is essential.1,2
Fig. 1.
AP radiograph of the right shoulder. This projection provides a general overview of the shoulder joint and hence is typically included in any standard shoulder radiograph. It is useful for evaluation of fractures or frank dislocation.
Fig. 2.
Axillary view of a right shoulder. This projection is centered on the glenohumeral joint and is useful in detection of more subtle anterior or posterior dislocations.
Fig. 3.
AP projection of the right shoulder take in external rotation. This projection is useful in detection of any glenohumeral joint arthritis or humeral head fractures.
Fig. 4.
AP projection of the right shoulder taken in internal rotation. Note that the lesser tuberosity (yellow arrow) is seen en face. This projection allows evaluation for Hill-sachs deformity.
3. Subacromial Impingement
Subacromial Impingement (SI) is the most common form of shoulder impingement. It is a painful condition predominantly secondary to compression of the subacromial bursa or supraspinatus tendon between the greater tuberosity of the humerus and the coracoacromial arch. Suitable imaging and clinical assessments can lead to prompt diagnosis and treatment, potentially preventing development to a complete tear of the RC.3 SI can manifest radiographically as abnormalities that impose upon the supraspinatus and cause an indirect reduction in the subacromial space (Fig. 5).
Fig. 5.
AP radiograph. There is inferolateral tilt of the acromion (blue line) with osteophytes at the inferior surface of the ACJ (yellow arrow) causing reduction in subacromial space.
There are anatomical variations associated with a greater incidence of impingement symptoms. For example, a type three acromion where the anterior acromion is hooked inferiorly has been associated with an increased prevalence of RC tears.4 Os acromiale, an anatomical variant which secondary to failure of fusion of the acromion ossification centres, are typically well seen on axillary projections (Fig. 6). This is traditionally considered an incidental finding, however there have been reports more recently that it can be a cause of SI and subsequent RC tear.3,5
Fig. 6.
(a) Axillary projection of a right shoulder demonstrating a zig-zag lucency between the acromion and main body of scapula (yellow arrow), this is an os acromiale and should not be confused as a fracture which is typically a straight non-corticated lucency. (b) Coronal CT of the same patient’s right shoulder confirming an os acromiale. Note the absence of any fat stranding which also makes this less likely to be secondary to an acute fracture.
4. Rotator cuff tear
RC tear is one of the most common cause of shoulder pain, with the supraspinatus tendon being most commonly involved. Prevalence of RC tear increases with age and the most significant clinical findings include pain whilst raising the abducted arm (arc of pain sign).4 Although torn tendons are indistinguishable on radiographs, this modality remains utilised in the initial assessment allowing assessment of osseous relationships within the shoulder. RC rupture due to high energy trauma with bony involvement can be detected on radiography (Fig. 7). A true AP view is most useful in evaluating chronic cuff tears and may demonstrate a reduced acromiohumeral interval, a normal value sits between 8 and 12 mm on AP view. The absence of the supraspinatus tendon acting as a barrier when it is torn and the unopposed action of the deltoid muscle pulling the humerus superiorly are the foundations for a ‘high riding shoulder’6 with reduced acromiohumeral distance and humeral head migration (Fig. 8). Although a decreased measurement is indicative of RC disease, it is important to note that this value can be artificially reduced by technical factors, such as parallel errors introduced by X-ray beam being centered too low.7
Fig. 7.
AP radiograph of a left shoulder with an avulsion fracture of the greater tuberosity at the site of the supraspinatus attachment (yellow arrow).
Fig. 8.
AP radiograph of a right shoulder demonstrating superior migration of the humeral head with a reduction in the subacromial space (yellow arrow) in keeping with a ‘high riding’ shoulder.
RC arthropathy is a spectrum of disease that develops in a RC deficient shoulder. Development of a rupture may lead to cuff arthropathy which occurs due to the expanding tear of the tendon and ascending migration of the humeral head, as well as underlying glenohumeral joint arthrosis. The AP view may demonstrate sclerosis and spurring of the acromion with concavity at the inferolateral aspect. Responsive radiographic changes at the insertion site on the greater tuberosity may also be seen, including sclerosis and subchondral cyst formation. Gradually, glenohumeral joint osteophytes can develop as a counteractive response to preserve joint congruity.1,2,6
5. Calcific tendonitis
Calcific tendonitis is a common condition related to the pathologic deposition of calcium hydroxyapatite crystals within the tendons. In many cases, this calcification may be asymptomatic however it can also be an important cause of painful joints. The supraspinatus tendon is most commonly affected, followed by the infraspinatus, subscapularis and teres minor respectively8 (Fig. 9, Fig. 10). The typical point of calcification in the supraspinatus tendon is approximately 1 cm from its insertion on the greater tubercle of the humerus, which is a site of relative avascularity. These calcific deposits decrease the space between the RC and the acromion, subsequently leading to SI.9
Fig. 9.
AP radiograph. Curvilinear, homogenous calcific deposit projected over the supraspinatus tendon (yellow arrow).
Fig. 10.
Axillary radiograph. Calcific deposit projected anterior to the lesser tuberosity of the humeral head. This is a case of subscapularis calcific tendinosis (yellow arrow).
Accurately positioned radiographs are essential to diagnose calcific tendonitis. Tangential radiographs of the concerned cortex are necessary to identify and accurately characterise the calcifications, with even a few degrees of inaccuracy causing the calcification to be imperceptible due to superimposition of neighbouring bony structures.7 Routine radiographs used for diagnosis and follow-up of cuff calcifications are the AP, outlet, and axillary views, as they offer valuable evidence concerning position and morphology of the deposits.10 Calcific tendinitis progresses through numerous stages and has distinct radiographic and pathologic features. In general, chronic calcifications appear crowded with sharp boundaries and hyperdense; whereas acute calcification appears less crowded and are sometimes ‘cloudy’ or demonstrate a ‘cotton-ball’ appearance indicating its resorptive phase. Findings on MR can be alarming with aggressive osseous changes with soft-tissue oedema and mistaken for an aggressive process such as infection or malignancy – this is mainly observed when there is intra-osseous extension of the calcific focus into the humeral head with erosion of cortical margins. This is a potential pitfall and highlights the importance of obtaining the correct modality for the clinical question and reviewing corresponding radiographic images even if cross-sectional imaging has been performed10, 8, 9 (Fig. 10).
6. Non-RC pathologies
Shoulder pain is a common and disabling complaint responsible for approximately 16% of musculoskeletal (MSK) disorders.11 Although many shoulder complaints are caused by rotator cuff (RC) pathology, it is important to consider causes of non-traumatic pain also. The most significant differential to consider is an apical lung tumour (pancoast tumour) (Fig. 11) which can cause referred shoulder pain by encroaching on the nerves.12 Other differentials can include secondary metastasis, glenohumeral or acromioclavicular joint arthritis, rheumatoid arthritis and post-traumatic osteolysis (Fig. 12, Fig. 13, Fig. 14, Fig. 15, Fig. 16). When assessing shoulder pain, it is important that clinicians maintain a high index of suspicion even in the most banal of conditions.
Fig. 11.
AP radiograph of the left shoulder obtained for shoulder pain. Note the large soft tissue opacity projected over the left lung apex (yellow arrow) representing a pancoast tumour.
Fig. 12.
AP radiograph. Subtle sclerotic lesion in the metaphysis of the proximal humerus (yellow arrow). Metastatic bone disease in a patient with known prostate carcinoma.
Fig. 13.
Modified axillary radiograph. Significant reduction of the glenohumeral joint space with subchondral sclerosis (yellow arrows). Osteophyte formation is noted at the inferior articular margin of the humerus (green arrow).
Fig. 14.
AP radiograph. Moderate reduction of the acromioclavicular space with subchondral sclerosis (yellow arrow) and cyst formation (green arrow).
Fig. 15.
AP radiograph. Large, peri-articular erosions of the humeral head. Further erosion of the distal clavicle with widening at the ACJ. Advanced rheumatoid disease.
Fig. 16.
AP radiograph. Cortical irregularity with erosion of the distal clavicle (yellow circle). Post-traumatic distal clavicle osteolysis. Note there is also lateral ‘down-sloping’ of the acromion.
Radiography remains the first line of imaging used in assessment of shoulder pain. Although not all causes of shoulder pain are evident on radiographs, this modality has its advantages due to its availability, speed and low cost. Though it carries some radiation exposure, it is non-invasive and can be interpreted widely among a multidisciplinary team.7
7. Ultrasound of the shoulder
7.1. Technique
The shoulder US examination is performed using high-frequency (9–15 MHz) linear broad-band array transducers. Each structure is evaluated in its longitudinal and transverse axes. Gentle heel-toe manoeuvres along the long axis and toggling movements along the transverse axis are done to avoid anisotropy which is a potential pitfall causing overdiagnosis of RC tear.
7.2. Sonographic appearances of normal structures
To be able to recognise an abnormality, one should familiarise themselves with the normal appearances of the area of interest. The typical normal sonographic appearances of different structures in the shoulder joint are detailed in Fig. 17. The advantages and disadvantages of ultrasound are discussed in Table 1.
Fig. 17.
US image of a normal supraspinatus (SST) tendon which typically appears hyperechoic with fibrillar echo patterns. Note the curvilinear hypoechogenicity within the tendon secondary to anisotropy and should not be confused for a tear. The humeral head bony cortex appears echogenic with posterior acoustic shadowing (yellow arrowheads). The overlying hyaline cartilage appears uniformly hypoechoic (green arrow). The overlying deltoid muscle typically appears relatively hypoechoic to normal tendons.
Table 1.
Advantages and disadvantages of ultrasound.
| Advantages | Disadvantages |
|---|---|
| Portability | Operator skill dependent |
| No radiation risk | Limited assessment of deeper structures, eg: labrum and cartilage |
| Dynamic assessment and allows real time comparison with the other side | Cannot detect intraosseous abnormality |
| Low cost | |
| Alternative to patients with contraindication for MRI or claustrophobia |
7.3. Sensitivities and specificities of US
A full-thickness rotator cuff tear can be diagnosed with 92.3% sensitivity and 94.4% specificity by using US, accuracy is slightly lower with diagnosing partial-thickness tear with 66.7% sensitivity and 93.5% specificity.13 When performed by experienced individuals, US and MR imaging have equivalent high sensitivity and specificity in the diagnosis of full- and partial-thickness RC tears.14,15 A meta-analysis by de Jesus et al.16 found that although MR arthrography was more accurate than both US and MR in diagnosis of RC tears, there was no significant difference between US and MR in diagnosis of RC tears.
7.4. Rotator cuff pathologies identified on US
Common rotator pathologies identified on ultrasound typically include calcific tendonitis, tendinopathy, subacromial-subdeltoid (SASD) bursitis and partial thickness tear.17 It is common for one to have a mixture of these pathologies.
7.5. Rotator cuff tears
The anterior aspect of the distal supraspinatus close to the rotator interval is a common site of tear; a tear more posteriorly near the conjoint insertion of SST and IST has been described with degenerative tears.18,19
7.6. Full thickness rotator cuff tears (FTRCT)
Most RC tears involve the SST and occur approximately 13–17 mm posterior to the LHBT, near the confluence of the SST and IST.19 FTRCT on US typically appear hypoechoic or anechoic representing a defect extending through the entire tendon, from the humeral articular surface to the SASD bursa (Fig. 18).
Fig. 18.
US image of a supraspinatus (SST) tendon with full-thickness tear at the footprint (yellow arrow) with retraction (blue arrow) to the level of the humeral head.
Secondary indirect signs of FTRCT include cortical irregularity of the greater tuberosity footprint, “cartilage interface” sign, glenohumeral joint effusion, herniation of the SASD bursa and deltoid muscle and SASD bursal effusion.20, 21, 22 The “cartilage interface sign” represents the accentuation of the hyperechoic interface between the hypoechoic torn tendon and the hypoechoic humeral head hyaline cartilage.20 Cortical irregularity and joint effusion have the highest sensitivity, specificity, positive and negative predictive values for the detection of full-thickness SST tears on US20. Findings of both glenohumeral joint effusion and SASD fluid suggest RC tear with a PPV of 95%.20 Massive tears are tears greater than 5 cm in width and/or involving two or more tendons (Fig. 19). Acute FTRCT commonly involve the mid substance of the tendon and are associated with joint or bursal effusion. Chronic FTRCT are more commonly associated with tendon retraction or non-visualization and less commonly associated with joint or bursal effusion.23
Fig. 19.
US image of the humeral head in sagittal plane. No tendon is seen at the expected attachment site of supraspinatus (yellow arrow) and infraspinatus (green arrow) tendons. There is depression of the overlying deltoid muscle into the tendon defect (white arrows).
7.7. Partial thickness rotator cuff tears (PTRCT)
PTRCT can appear as focal hypoechogenicity involving the articular or bursal surface (Fig. 20) or as mixed echogenicity within the critical zone of the tendon disrupting the fibrillar pattern (interstitial delaminating tears). Articular sided tears are more common in young <40years and involve SST anteriorly and distally at the greater trochanter.18,23 Cartilage interface sign is more specific for articular sided partial thickness tear. Herniation of the SASD bursa and deltoid muscle into the tendon defect is seen more commonly in bursal-sided partial thickness tears. Bursal-sided tears are more frequently associated with retraction and volume loss. Secondary findings of partial-thickness RC tears include cortical irregularity or “pitting” at the tendon insertion and transducer pressure–induced depression of the SASD bursa into the tear defect, which is seen with bursal sided tears. An acute tear usually lacks cortical irregularity.24
Fig. 20.
US image of a supraspinatus tendon with partial thickness tear at its footprint (yellow arrow) extending to the articular surface. Note that the bursal surface remains intact (red arrowheads).
7.8. Tendinopathy
Tendinopathy is an umbrella term for abnormalities within the tendon including tendinosis and tenosynovitis. Tendinosis represents the background chronic changes within a tendon (usually degenerative); tenosynovitis represents acute inflammation of the synovial sheath covering the tendon. However, most people tend to use the term tendinopathy and tendinosis interchangeably. Tendinosis typically manifests as a heterogenous ill-defined tendon25 and can be associated with increased tendon thickness and/or cortical irregularity (Fig. 21). RC tears often co-exist with underlying tendinosis.
Fig. 21.
US image of a thickened supraspinatus tendon with heterogenous echogenicity throughout consistent to moderate tendinosis. There is no tear within this tendon.
7.9. Calcific tendinitis
RC calcifications appear as fluffy or well-defined hyper-echogenicities within the tendon, usually with posterior acoustic shadowing depending on its calcium content (Fig. 22). Increased colour doppler signal is associated with the resorptive phase.26 SST is the most commonly affected tendon and the calcific foci can sometimes migrate or extend into the subacromial subdeltoid bursa. Hydroxyapatite deposition typically occurs approximately 10 mm from the SST insertion on the greater tuberosity.
Fig. 22.
US image demonstrating amorphous echogenicities (yellow arrows) with posterior acoustic shadowing within the supraspinatus tendon representing calcific deposit within the tendon.
7.10. RC atrophy
On US, fatty infiltration and atrophy appear as increased echogenicity of the muscle with resultant poor differentiation between tendon and muscle and reduced muscle bulk. SST atrophy is associated with anterior RC tears and IST atrophy with size of full thickness tears. Taking teres minor as reference IST atrophy defined as its area less than 50% of TM. Teres minor and subscapularis atrophy are not as commonly seen.
7.11. Tenosynovitis
Tenosynovitis can manifest as anechoic fluid (out of proportion to joint fluid) or hypoechoic synovial hypertrophy surrounding the tendon with associated debris or echogenic synovial hypertrophy (Fig. 23). On power Doppler US images, increased flow within an abnormally thickened tendon sheath represents synovial proliferation.
Fig. 23.
(a) US image of the long head of biceps tendon (blue arrow) in transverse axis demonstrating moderate biceps tendon sheath effusion (yellow). (b) US image of the long head of biceps tendon in longitudinal axis (blue arrow) with increased Doppler flow signal indicating acute tenosynovitis.
7.12. Biceps subluxation and dislocation
When not visualised in the bicipital groove, the long head of biceps (LHB) tendon may have displaced into various positions. For example, superficial or medial to the lesser tuberosity, into a SSC tendon tear or through a SSC tear into the glenohumeral joint.
Tear of the bicipital reflection pulley which consists of SGHL and CHL in rotator interval, predispose the tendon to instability. Chondral print sign-subchondral bone cortex irregularity adjacent to BT in rotator interval is an indirect sign of instability.
7.13. Guided injections
Ultrasound is increasingly used for various diagnostic and therapeutic procedures as it is without risk of radiation and fairly easy to access. Commonly performed diagnostic US-guided procedures include soft tissue biopsy, aspiration of joint fluid or collection. Some centres use ultrasound instead of fluoroscopy to inject contrast into joints for arthrogram. Ultrasound-guided steroid injections into the joints, bursa (Fig. 24) or tendon sheath can also be performed in various parts of the body for both diagnostic and therapeutic purposes. In patients with calcific tendonitis, ultrasound-guided barbotage can help to break up the calcification and speed up the resorptive phase; steroid injection into the subacromial bursa is commonly performed during barbotage. Ultrasound can also be used for tendon fenestration±plasma rich protein injection to facilitate healing of partial tendon tear.
Fig. 24.
(a) US image demonstrating a thickened subacromial bursa (blue arrow). (b) US-guided injection of the subacromial bursa. The needle appears as a linear hyperechoic structure (red arrows) with its tip within the subacromial bursa which is now distended with injectate fluid (yellow arrow) usually consisting of steroid and local anaesthetic mixture.
8. Magnetic resonance imaging
8.1. Imaging protocol
Plain magnetic resonance (MR) imaging provides superior soft tissue contrast and resolution making it a useful tool in assessment of the shoulder for rotator cuff pathology. Every institution has their own imaging protocol but an example protocol of a plain MR and MR arthrogram (MRA) of the shoulder as recommended by the European Society of Skeletal Radiology27 are detailed in Table 2, Table 3.
Table 2.
Example protocol of plain MR shoulder from ESSR.27
| Plane | Sequence | FOV (max) | Slice thickness | TE | Matrix (min) |
|---|---|---|---|---|---|
| Axial | Intermediate-FS | 16 cm | 3.5 mm | 40–60 | 256 × 256 |
| Coronal Oblique | Intermediate-FS | 16 cm | 3.5 mm | 40–60 | 256 × 256 |
| Coronal Oblique | T2 | 16 cm | 3.5 mm | 40–60 | 256 × 256 |
| Sagittal Oblique | Intermediate-FS | 16 cm | 3.5 mm | 40–60 | 256 × 256 |
| Sagittal Oblique | T1 | 16 cm | 4 mm | Min | 256 × 256 |
| ∗(Optional) Axial | GRE | 16 cm | 3.5 mm | 10–20 | 256 × 256 |
FOV = field of view FS = fat-saturated GRE = gradient echo.
Table 3.
Example protocol of an MRA of the shoulder from ESSR.27
| Plane | Sequence | FOV (max) | Slice thickness | TE | Matrix (min) |
|---|---|---|---|---|---|
| Axial | T1 | 16 cm | 3.5 mm | Min | 256 × 256 |
| Axial | PD-FS | 16 cm | 3.5 mm | 35–45 | 256 × 256 |
| Coronal Oblique | T1-FS | 16 cm | 3.5 mm | Min | 256 × 256 |
| Coronal Oblique | T2-FS | 16 cm | 3.5 mm | 80–100 | 256 × 256 |
| Sagittal Oblique | PD-FS | 16 cm | 3.5 mm | 40–60 | 256 × 256 |
| ∗(Optional) ABER | T1-FS | 16 cm | 3.5 mm | Min | 256 × 256 |
FOV = field of view PD = proton density FS = fat-saturated.
Some institutions may add an ABER T1-FS view onto their MRA protocol where the patient’s shoulder is in abduction and external rotation with their hands above and behind their head. This positioning stretches the inferior glenohumeral ligament which in turn improves detection rate of anteroinferior labroligamentous lesions28 and other labral and rotator cuff tears.29 However, there may be some practical challenges in setting up the sequence in a closed-bore MR scanner30 and unfamiliarity with interpretation of the alignment of structures on this view.
8.2. Plain MR vs MRA
Plain MR of the shoulder typically takes about 25 min. MR arthrogram of the shoulder is an MR examination after arthrogram has been performed. An arthrogram involves contrast injection into the shoulder joint to outline the intra-articular structures. This can be done under fluoroscopy guidance and/or ultrasound depending on local institutional facilities. It carries its own risk as an invasive procedure albeit small with possible ionizing radiation exposure if done under fluoroscopic guidance. It also requires radiologist availability to perform the procedure and coordination with MR for subsequent scan slot ideally within 1 h of the injection. Fig. 25, Fig. 26, Fig. 27 depict the differences between MRI and MRA.
Fig. 25.
Axial PD-FS plain MR of the right shoulder without intra-articular contrast administration at the level of the subscapularis tendon. Note that the structures are in close proximity to each other in the absence of any effusion or contrast administration to distend the joint.
Fig. 26.
Fluoroscopic spot image of a right shoulder arthrogram. Note the iodine-based contrast has flown into the glenohumeral joint (yellow arrow) confirming correct intra-articular needle position.
Fig. 27.
Axial T1-FS MR Arthrogram of the left shoulder of a different patient. Note that the joint capsule is distended with contrast which helps separate structures in the area to aid with interpretation.
A meta-analysis by JS Roy et al.31 reported similar rate of detection of full-thickness tear on MRI vs MRA with pooled sensitivity of 0.90 for both MRI and MRA; pooled specificity was 0.93 for MRI and 0.95 for MRA. However, the authors had reported that MRA was superior in detecting partial-thickness tear - pooled sensitivity was 0.67 on MRI and 0.83 on MRA; pooled specificity were similar at 0.94 on MRI and 0.93 on MRA. Plain MR is comparable to MRA in detection of full-thickness rotator cuff tear. However, plain MR is not as sensitive as MRA in detection of partial-thickness tear, although its specificity is similar to MRA. In conjunction with the risks and additional resources required highlighted above, a plain MR is therefore recommended in the first instance rather than MRA,32 particularly when the patient is unlikely to be a surgical candidate.
8.3. Subacromial impingement
The supraspinatus tendon glides under the coracoacromial arch and can be impinged when the subacromial space is reduced (Fig. 28a). The configuration of the acromion process is thought to play an important role. Bigliani’s classification system is commonly used, it divides the acromial configuration into three types. Type 3 acromion (hooked type) is thought to be related to increased incidence of rotator cuff tears.33 Later studies34, 35, 36 have suggested that parameters such as lateral acromial angle, acromial index or critical shoulder angle are more reliable indicators compared to Bigliani type classification. Acromial osteophyte or enthesophyte (Fig. 28b) have been reported to be strongly associated with FTRTCs.36 A brief description of the acromial configuration and any acromial enthesophyte is important as the clinical team could potentially perform bone shaving during arthroscopy or surgical repair of the tendons.
Fig. 28.
Plain shoulder MRI of different patients. (a) Coronal PD-FS MR a right shoulder demonstrating mild bursal thickening and excess fluid in the subacromial bursa (yellow arrows) representing subacromial bursitis. Note mild to moderate underlying supraspinatus tendinosis, no tear. (b) Sagittal T1 image of another shoulder demonstrating a prominent inferior acromial enthesophyte (red arrow) indenting onto the supraspinatus myotendinous junction (green curve line) leading to reduced subacromial space and impingement.
8.4. Tendinosis
Rotator cuff tendinosis refers to chronic degenerative changes to the tendon where there is loss of normal collagen matrix organisation, myxoid degeneration and fibrocartilaginous changes histologically.37 The normal rotator cuff tendon should demonstrate a homogenous low signal on all MR sequences. Instead, when there is rotator cuff tendinosis, the tendon may appear thickened with intrasubstance heterogenous intermediate signal (Fig. 29), often there is associated bony irregular changes or even subchondral cystic changes as well. Note that the intermediate signal seen in tendinosis is not as high (bright) as seen in rotator cuff tear, both entities often exist together.
Fig. 29.
Plain shoulder MR of different patients. (a) Coronal PD-FS MR a right shoulder demonstrating intermediate signal in the distal supraspinatus tendon (yellow circle) representing mild tendinosis without cuff tear. Note that the signal is not as high as seen in a rotator cuff tear. (b) Axial T2-FS of a left shoulder demonstrating a thickened subscapularis with intermediate signal in keeping with moderate tendinosis without tear (orange circle).
8.5. Cuff tear – partial thickness rotator cuff tear
Partial-thickness rotator cuff tears (PTRCTs) can be classified according to its location: articular, bursal or interstitial. Articular sided tears are more common (Fig. 30a). Bursal sided tears (Fig. 30b and c) are not usually detected on standard approach arthroscopy and are repaired via a bursoscopy, hence it is important to detect this on pre-op imaging. The PTRCT’s size should ideally be measured in three dimensions and is graded according to percentage of tendon thickness involved38: Grade 1 (<25%), grade 2 (25–50%) and grade 3 (>50%). Low grade PTRCT (grade 1 or 2) may initially be managed conservatively whilst high grade PTRCT (grade 3) may warrant surgical repair if clinically appropriate. Other important points that should be included in the report include shape of the tear (crescent, U or L-shape), any extension of tear into other structures or tendons as well as rotator cuff muscle bulk. All of the above points in conjunction with patient’s clinical status contribute to the clinical team’s decision to operate or not as well as choice of surgical approach.
Fig. 30.
(a). Coronal T2 FS plain MR of the right shoulder demonstrates curvilinear high signal at the articular side of the supraspinatus tendon medially involving between 25 and 50% of tendon thickness, in keeping with an articular-sided grade 2 partial thickness tear of the supraspinatus tendon.
Fig. 30. (b) and (c) Coronal and Sagittal T2-FS plain MR of the left shoulder of a different patient demonstrates a curvilinear high signal at the bursal side of the mid-portion of the supraspinatus tendon (blue arrow) involving approximately 50% of tendon thickness, in keeping with a high grade bursal-sided partial thickness tear of the supraspinatus tendon.
8.6. Cuff tear – full thickness rotator cuff tear
Full thickness rotator cuff tears (FTRCTs) are tears involving 100% of the tendon thickness. It can be classified according to its size: small (<1 cm), medium (1–3 cm), large (3–5 cm) or massive (>5 cm). The larger the tear, the less favorable the outcome. Size of the tear may also influence the choice of surgical approach. Presence of tendon retraction should also be noted and is graded according to its location whether near the insertion site, at level of humeral head, at the level of glenoid or medial to it, see Fig. 31.
Fig. 31.
(a) and (b) Coronal and sagittal T2-FS MR of the right shoulder demonstrates high signal throughout the distal supraspinatus tendon without any appreciable tendon fibre representing a full-thickness tear (rupture) of the supraspinatus tendon. Note the tendon retraction (blue arrow) to the level of the humeral head. Note the loss of normal convex contour of the supraspinatus tendon representing full-thickness tear (green arrow). There is also high signal extending posteriorly into the infraspinatus tendon signifying tear extension into the infraspinatus tendon (yellow arrow).
The rotator cuff muscle bulk should also be examined for atrophy and/or fatty degeneration as the rotator cuff muscle quality would also influence the surgical technique and outcome. The degree of volume loss can be graded according to occupation ratio39 (Table 4). This is calculated using the ratio of supraspinatus muscle belly surface area to the supraspinatus fossa surface area on sagittal oblique imaging (Y-shaped view).
Table 4.
Classification of supraspinatus belly atrophy based on occupation ratio of the supraspinatus fossa as proposed by Thomazeau et al.39
| Degree of volume loss | Description | Occupation ratio |
|---|---|---|
| Stage 1 | Normal or slightly atrophied | 0.6–1.0 |
| Stage 2 | Moderate atrophy | 0.4–0.6 |
| Stage 3 | Serious or severe atrophy | Less than 0.4 |
[Insert Table 4. Classification of supraspinatus belly atrophy based on occupation ratio of the supraspinatus fossa as proposed by Thomazeau et al.39].
Alternatively, some use the tangent sign to quantify the degree of supraspinatus belly atrophy. For this, a line is drawn between the superior border of the scapular spine and superior border of the medial coracoid process – a normal supraspinatus muscle belly would lie above this line (Fig. 32a and b) whilst an atrophied supraspinatus would lie below this line (Fig. 32c and d). On the other hand, the degree of fatty degeneration should also be noted and can be graded according to Goutallier classification40 (Table 5), also assessed on a standard T1 sagittal oblique sequence. The higher the degree of rotator cuff muscle fatty degeneration, the higher the risk of post-surgical tear recurrence.41
Fig. 32.
(a) and (b) Sagittal T1 MR of the shoulder through the medial border of the coracoid process (Y-shaped view) demonstrating a normal supraspinatus muscle belly bulk lying above the green line drawn between the superior border of the scapular spine and medial coracoid (negative tangent sign). The occupation ratio is calculated by dividing the surface area of the supraspinatus muscle belly (blue shaded area) by the supraspinatus fossa surface area (yellow shaded area). The occupation ratio here is normal (>0.6).
Fig. 32. (c) and (d) Sagittal T1 MR of another patient’s shoulder demonstrating supraspinatus atrophy. The supraspinatus belly is below the green line drawn between the superior border of the scapular spine and medial coracoid (positive tangent sign). Note the amount of fatty replacement within the supraspinatus belly (green arrow) representing at least a Goutallier stage 3 fatty degeneration. The occupation ratio here is < 0.4 indicating stage 3 (severe) atrophy. Note that there is also mild infraspinatus atrophy and stage 2 fatty degeneration (yellow arrow).
Table 5.
Goutallier classification of rotator cuff muscle fatty degeneration 41
| Stage 0 | Normal muscle |
|---|---|
| Stage 1 | Some fatty streaks |
| Stage 2 | Less than 50% fatty muscle atrophy |
| Stage 3 | 50% fatty muscle atrophy |
| Stage 4 | Greater than 50% fatty muscle atrophy |
8.7. Calcific tendonitis
Calcific tendonitis occurs when there is deposition of calcium hydroxyapatite crystals within the tendons, usually in the supraspinatus tendon within 1 cm from the insertional site. Radiography and ultrasonography remain the most useful modality in detecting calcific tendonitis (Fig. 33) as the calcific deposits may be too small to appreciate on MR. MR can however exclude any other pathology that could contribute to the patient’s symptoms. Findings on MR can be alarming with intra-osseous extension of the calcific focus into the humeral head resulting in cortical erosions and associated soft tissue oedema secondary to the calcific tendonitis. This can mimic more aggressive pathologies such as malignancy or infection. Therefore, correlation with radiography is crucial in avoiding this pitfall.
Fig. 33.
(a) and (b) Coronal and sagittal PD-FS MR of the left shoulder demonstrating a low signal body (yellow arrow) within the posterior supraspinatus tendon near the conjoint tendon. On MR alone, appearances are non-specific and can represent underlying tendinosis or calcification. (c) Corresponding AP radiograph of the left shoulder of the same patient confirming calcification within the supraspinatus tendon (blue arrow). This case highlights importance of correlating MR findings with radiographic features.
9. Conclusion
All three modalities (radiography, ultrasound and MRI) each have a role in assessment of the rotator cuff. This paper outlines the advantages and disadvantages of each modality and illustrates typical radiological findings of common rotator cuff pathologies in each modality.
References
- 1.Chang E.Y., Chung C.B. Imaging diagnosis of rotator cuff pathology and impingement syndromes. In: Bencardino J.T., editor. The Shoulder- Imaging Diagnosis with Clinical Implications. Springer; Switzerland: 2019. pp. 87–127. [Google Scholar]
- 2.Giannini S., Ondo E.P.A., Sabino G. Instrumental evaluation: X-rays, MRI. In: Gumina S., editor. Rotator Cuff Tear-Pathogenesis, Evaluation and Treatment. Springer; Switzerland: 2017. pp. 169–185. [Google Scholar]
- 3.Garving C., Jakob S., Bauer I., Nadjar R., Brunner U. Impingement syndrome of the shoulder. Deutsches Aerzteblatt Online. 2017 doi: 10.3238/arztebl.2017.0765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Itoi E. Rotator cuff tear: physical examination and conservative treatment. J Orthop Sci. 2013;18(2):197–204. doi: 10.1007/s00776-012-0345-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.You T., Frostick S., Zhang W., Yin Q. Os acromiale: reviews and current perspectives. Orthop Surg. 2019;11(5):738–744. doi: 10.1111/os.12518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lee S., Chun K., Lee S. Relationships between rotator cuff tear types and radiographic abnormalities. Journal of the Korean Society of Radiology. 2014;71(5):258. doi: 10.3348/jksr.2014.71.5.258. [DOI] [Google Scholar]
- 7.Bianchi S., Prato N., Martinolli C., Derchi L.E. Shoulder radiography. In: Davies A.M., Hodler J., editors. Imaging of the Shoulder Techniques and Applications. Springer; New York: 2005. pp. 3–15. [Google Scholar]
- 8.Sansone V., Maiorano E., Galluzzo A., Pascale V. Calcific tendinopathy of the shoulder: clinical perspectives into the mechanisms, pathogenesis, and treatment. Orthop Res Rev. 2018;10:63–72. doi: 10.2147/orr.s138225. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Marques I., Ressurreição J., Soares J.T. Poster Presentation at: European Congress of Radiology. 2014 March 6-10. Imaging of calcific tendinitis of the shoulder: a multimodality approach. Vienna. Available from: [DOI] [Google Scholar]
- 10.Chianca V., Albano D., Messina C. Rotator cuff calcific tendinopathy: from diagnosis to treatment. Acta Biomed. 2018;89(1-S):186–196. doi: 10.23750/abm.v89i1-S.7022. Published 2018 Jan 19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Elaassar M., Emadeldin A., Zidan D., Sakr H., Abdelazim H. Chronic non-traumatic shoulder pain: evaluation with ultrasonography and conventional mr imaging. QJM: Int J Med. 2020;(Supplement_1):113. doi: 10.1093/qjmed/hcaa068.013. [DOI] [Google Scholar]
- 12.Al Shammari M., Hassan A., Al Jawad M. Pancoast tumor: the overlooked etiology of shoulder pain in smokers. American Journal of Case Reports. 2020;21 doi: 10.12659/ajcr.926643. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bicos J. Biomechanics and anatomy of the proximal biceps tendon. Sports Med Arthrosc Rev. 2008;16(3):111–117. doi: 10.1097/JSA.0b013e31818247a1. [DOI] [PubMed] [Google Scholar]
- 14.Fotiadou A.N., Vlychou M., Papadopoulos P., Karataglis D.S., Pal- ladas P., Fezoulidis I.V. Ultrasonography of symptomatic rotator cuff tears compared with MR imaging and surgery. Eur J Radiol. 2008;68(1):174–179. doi: 10.1016/j.ejrad.2007.11.002. [DOI] [PubMed] [Google Scholar]
- 15.Teefey S.A., Rubin D.A., Middleton W.D., Hildebolt C.F., Leibold R.A., Yamaguchi K. Detection and quantification of rotator cuff tears: comparison of ultrasonographic, magnetic resonance imag- ing, and arthroscopic findings in seventy-one consecutive cases. J Bone Joint Surg Am. 2004;86-A(4):708–716. [PubMed] [Google Scholar]
- 16.de Jesus J.O., Parker L., Frangos A.J., Nazarian L.N. Accuracy of MRI, MR arthrography, and ultrasound in the diagnosis of rotator cuff tears: a meta-analysis. AJR Am J Roentgenol. 2009;192(6):1701–1707. doi: 10.2214/AJR.08.1241. [DOI] [PubMed] [Google Scholar]
- 17.Zheng F., Wang H., Gong H., Fan H., Zhang K., Du L. Role of ultrasound in the detection of rotator-cuff syndrome: an observational study. Med Sci Mon Int Med J Exp Clin Res. 2019;25:5856–5863. doi: 10.12659/MSM.915547. Published 2019 Aug 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Schaeffeler C., Mueller D., Kirchhoff C., Wolf P., Rummeny E.J., Woertler K. Tears at the rotator cuff footprint: prevalence and imaging characteristics in 305 MR arthrograms of the shoulder. Eur Radiol. 2011;21(7):1477–1484. doi: 10.1007/s00330-011-2066-x. [DOI] [PubMed] [Google Scholar]
- 19.Kim H.M., Dahiya N., Teefey S.A. Location and initiation of degenerative rotator cuff tears: an analysis of three hundred and sixty shoulders. J Bone Joint Surg Am. 2010;92(5):1088–1096. doi: 10.2106/JBJS.I.00686. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Jacobson J.A., Lancaster S., Prasad A., van Holsbeeck M.T., Craig J.G., Kolowich P. Full-thickness and partial-thickness supraspinatus tendon tears: value of US signs in diagnosis. Radiology. 2004;230(1):234–242. doi: 10.1148/radiol.2301020418. [DOI] [PubMed] [Google Scholar]
- 21.Wohlwend J.R., van Holsbeeck M., Craig J. The association between irregular greater tuberosities and rotator cuff tears: a sonographic study. AJR Am J Roentgenol. 1998;171(1):229–233. doi: 10.2214/ajr.171.1.9648794. [DOI] [PubMed] [Google Scholar]
- 22.Hollister M.S., Mack L.A., Patten R.M., Winter T.C., 3rd, Matsen F.A., 3rd, Veith R.R. Association of sonographically detected subacromial/subdeltoid bursal effusion and intraarticular fluid with rotator cuff tear. AJR Am J Roentgenol. 1995;165(3):605–608. doi: 10.2214/ajr.165.3.7645478. [DOI] [PubMed] [Google Scholar]
- 23.Teefey S.A., Middleton W.D., Bauer G.S., Hildebolt C.F., Yamaguchi K. Sonographic differences in the appearance of acute and chronic full-thickness rotator cuff tears. J Ultrasound Med. 2000;19(6):377–383. doi: 10.7863/jum.2000.19.6.377. [DOI] [PubMed] [Google Scholar]
- 24.Mochizuki T., Sugaya H., Uomizu M. Humeral insertion of the supraspinatus and infraspinatus. New anatomical findings regarding the footprint of the rotator cuff. J Bone Joint Surg Am. 2008;90(5):962–969. doi: 10.2106/JBJS.G.00427. [DOI] [PubMed] [Google Scholar]
- 25.Bachmann G.F., Melzer C., Heinrichs C.M., Möhring B., Rominger M.B. Diagnosis of rotator cuff lesions: comparison of US and MRI on 38 joint specimens. Eur Radiol. 1997;7(2):192–197. doi: 10.1007/s003300050133. [DOI] [PubMed] [Google Scholar]
- 26.C Chiou H.J., Chou Y.H., Wu J.J., Hsu C.C., Huang D.Y., Chang C.Y. Evaluation of calcific tendonitis of the rotator cuff: role of color Doppler ultrasonography. J Ultrasound Med. 2002;21(3):289–297. doi: 10.7863/jum.2002.21.3.289. [DOI] [PubMed] [Google Scholar]
- 27.European Society of Skeletal Radiology Sports Sub-committee . ESSR; 2016. Guidelines for MR Imaging of Sports Injuries. [Google Scholar]
- 28.Tian C., Cui G., Zheng Z., Ren A. The added value of ABER position for the detection and classification of anteroinferior labroligamentous lesions in MR arthrography of the shoulder. Eur J Radiol. 2013;82(4):651–657. doi: 10.1016/j.ejrad.2012.11.038. [DOI] [PubMed] [Google Scholar]
- 29.Saleem A., Lee J., Novak L. Usefulness of the abduction and external rotation views in shoulder MR arthrography. Am J Roentgenol. 2008;191(4):1024–1030. doi: 10.2214/ajr.07.3962. [DOI] [PubMed] [Google Scholar]
- 30.Aydingoz U., Maras Ozdemir Z., Ergen F. Demystifying ABER (ABduction and External Rotation) sequence in shoulder MR arthrography. Diagn Interventional Radiol. 2014;20(6):507–509. doi: 10.5152/dir.2014.14117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Roy J., Braën C., Leblond J. Diagnostic accuracy of ultrasonography, MRI and MR arthrography in the characterisation of rotator cuff disorders: a systematic review and meta-analysis. Br J Sports Med. 2015;49(20):1316–1328. doi: 10.1136/bjsports-2014-094148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Liu F., Cheng X., Dong J., Zhou D., Han S., Yang Y. Comparison of MRI and MRA for the diagnosis of rotator cuff tears. Medicine (Baltim) 2020;99(12) doi: 10.1097/md.0000000000019579. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Panni A., Milano G., Lucania L., Fabbriciani C., Logroscino C. Histological analysis of the coracoacromial arch: correlation between age-related changes and rotator cuff tears. Arthrosc J Arthrosc Relat Surg. 1996;12(5):531–540. doi: 10.1016/s0749-8063(96)90190-5. [DOI] [PubMed] [Google Scholar]
- 34.Balke M., Schmidt C., Dedy N., Banerjee M., Bouillon B., Liem D. Correlation of acromial morphology with impingement syndrome and rotator cuff tears. Acta Orthop. 2013;84(2):178–183. doi: 10.3109/17453674.2013.773413. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Moor B., Wieser K., Slankamenac K., Gerber C., Bouaicha S. Relationship of individual scapular anatomy and degenerative rotator cuff tears. J Shoulder Elbow Surg. 2014;23(4):536–541. doi: 10.1016/j.jse.2013.11.008. [DOI] [PubMed] [Google Scholar]
- 36.Pandey V., Vijayan D., Tapashetti S. Does scapular morphology affect the integrity of the rotator cuff? J Shoulder Elbow Surg. 2016;25(3):413–421. doi: 10.1016/j.jse.2015.09.016. [DOI] [PubMed] [Google Scholar]
- 37.Dean B., Franklin S., Carr A. A systematic review of the histological and molecular changes in rotator cuff disease. Bone Joint Res. 2012;1(7):158–166. doi: 10.1302/2046-3758.17.2000115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Matthewson G., Beach C., Nelson A. Partial thickness rotator cuff tears: current concepts. Adv Orthop. 2015;2015:1–11. doi: 10.1155/2015/458786. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Thomazeau H., Rolland Y., Lucas C., Duval J., Langlais F. Atrophy of the supraspinatus belly Assessment by MRI in 55 patients with rotator cuff pathology. Acta Orthop Scand. 1996;67(3):264–268. doi: 10.3109/17453679608994685. [DOI] [PubMed] [Google Scholar]
- 40.Goutallier D., Postel J.M., Bernageau J., Lavau L., Voisin M.C. Fatty muscle degeneration in cuff ruptures. Pre- and postoperative evaluation by CT scan. Clin Orthop Relat Res. 1994;304:78–83. [PubMed] [Google Scholar]
- 41.Goutallier D., Postel J., Gleyze P., Leguilloux P., Van Driessche S. Influence of cuff muscle fatty degeneration on anatomic and functional outcomes after simple suture of full-thickness tears. J Shoulder Elbow Surg. 2003;12(6):550–554. doi: 10.1016/s1058-2746(03)00211-8. [DOI] [PubMed] [Google Scholar]