Musculoskeletal ultrasound has expanded opportunities in the field of diagnostic ultrasound. The modality is readily available, economical, and portable. Its real-time capability helps render clinical correlation of the site of pain and aids in comparison with the contralateral side. Movement of the tendons and joints can be directly visualized with dynamic ultrasound scanning. Unlike other applications, musculoskeletal ultrasound is usually1 not affected by body habitus, motion artifacts, or intervening structures such as bowel gas. Structures such as tendons are better visualized with ultrasound than with MRI. Tendons appear as signal void on MRI but show a characteristic internal architecture on ultrasound.

Shoulder ultrasound has been the most prominent application of musculoskeletal imaging, as it has been used to evaluate the rotator cuff since the mid '80s. During the early days, lower frequency transducers of 7.5 MHz were used. Combined with limited experience, this was probably the reason for the low reported sensitivity of around 70%. Advances in transducer technology with frequencies reaching 13 to 15 MHz have improved the near-field resolution considerably and given shoulder and musculoskeletal ultrasound a much-needed boost. Better correlation of ultrasound findings with arthroscopic and surgical findings and more standardized diagnostic criteria have also helped make shoulder ultrasound a credible diagnostic modality, with sensitivity better than 90%. Studies have shown sensitivity for full-thickness tears ranging from 94% to 100% and sensitivity for partial-thickness tears ranging from 93% to 95%,1,2 with specificity for both being 94%. A recent study showed sensitivity of 100% with an overall accuracy of 96%.3

Among musculoskeletal complaints, shoulder pain follows only the most common complaints of low back pain and carpal tunnel syndrome. Subacromial impingement with subsequent tendinitis and bursitis is frequently found in young adult patients, whereas rotator cuff tears are a common cause of shoulder pain in patients over age 40. Incidence of rotator cuff tears rises from 25% in patients in their 50s to 50% in patients in their 70s.4 Ultrasound provides a reliable noninvasive tool in differentiating complete rotator cuff tears from other pathologies, since the majority of subacromial impingement and incomplete rotator cuff tears may be successfully managed with conservative treatment.5

SHOULDER TECHNIQUE

Shoulder ultrasound is performed preferably with the patient sitting and facing the examiner. The images on the screen are oriented as if one is facing the patient (similar to an x-ray). When imaging any tendon, it is important to make sure that the transducer is perpendicular to the tendon structure. This will ensure that normal fibers appear echogenic. If the ultrasound beam that hits the tendon fibers is not perpendicular, the tendon may appear hypoechoic. This reading is called anisotropy, and the artifact may lead to an erroneous tendon diagnosis. Anisotropy can be corrected with heel-toeing the transducer to align the tendon fibers perpendicular to the transducer.

The rotator cuff consists of the following tendons: supraspinatus, subscapularis, infraspinatus, and teres minor. Other important structures to remember during shoulder ultrasound are the biceps tendon and the deltoid muscle.

- Biceps tendon is the first structure imaged. The examination is started with the shoulder in the neutral position so that the biceps tendon groove is directly anterior. The biceps tendon is examined in the axial plane at the level of the bicipital groove and throughout the length of the biceps tendon sheath. The bicipital groove appears as a concavity in the bony surface of the humerus, with the lesser tuberosity located medially and the greater tuberosity located laterally. The biceps tendon appears as a hyperechoic oval structure within the bicipital groove. On long axis, the tendon appears as hyperechoic parallel fibers apposed to the anterior surface of the humerus (Figure 1). A small amount of intra-articular fluid surrounding the biceps tendon can be normally seen.

- Subscapularis tendon can be identified by starting in the axial plane over the biceps tendon groove and sliding the transducer medially. The subscapularis tendon will then be visualized inserting onto the medial aspect of the lesser tuberosity. Passive internal and external rotation of the arm will help in assessing the integrity of the tendon. On transverse scan, the tendon appears oval and is helpful in detecting tendinous attrition in patients with chronic anterior dislocation.

- Supraspinatus tendon is examined with the patient resting his or her hand as if it were in the back pocket with the elbow rotated as posterior as possible. This will expose more of the supraspinatus tendon from under the acromion process so that it can be sonographically imaged. Small tears are also often accentuated with this maneuver. The tendon is visualized as a band of medium-level echoes deep to the subdeltoid bursa and superficial to the bright echoes originating from the greater tuberosity. Articular cartilage of the humeral head can be visualized as a hypoechoic band just superficial to the bony margin. In longitudinal axis, the supraspinatus tendon appears as a beak-shaped soft-tissue structure extending from under the acromion to its attachment along the greater tuberosity (Figure 2). With passive abduction and adduction with the patients palm parallel to the greater tuberosity, small effusion in the subdeltoid bursa can be visualized. The superficial aspect of the rotator cuff and its interface with the deep aspect of the deltoid muscle should have a convex normal anatomic finding because loss of this convexity is one of the important ways of identifying rotator cuff tears.

- Infraspinatus tendon as imaged from the posterior aspect is visualized as a beak-shaped soft-tissue structure that runs horizontally and progressively thins as it approaches its attachment to the posterior aspect of the greater tuberosity. Passive internal and external rotation is helpful in confirming the integrity of the tendon. Additional structures visualized from the posterior aspect are the posterior glenoid labrium, which appears as a hyperechoic, triangular structure deep to the tendon, and the articular cartilage of the humeral head, which is the thin, hypoechoic layer superficial to the high-level echoes originating from the bony surface.

- Teres minor tendon is the trapezoidal soft-tissue structure with oblique internal echoes.

Distinction between the supraspinatus and infraspinatus tendons near the insertion is not generally possible on ultrasound, but it is known that the supraspinatus tendon measures about 1.5 cm in width. Therefore, the first 1.5 cm of rotator cuff located posterior and lateral to the intra-articular portion of the biceps tendon represents supraspinatus tendon, and the cuff posterior to this represents infraspinatus tendon and teres minor tendon.

PATHOLOGY OF SHOULDER

Shoulder ultrasound primarily involves scanning for rotator cuff tears, which can have the following appearance:

-Full-thickness tears can be large or small. In large full-thickness tears, the rotator cuff tears completely and the fragments retract from each other. Hence, no cuff is visualized. This creates a defect, which if filled with the joint fluid will appear as a hypoechoic or anechoic discontinuity in the rotator cuff (Figure 3). Absence of the cuff tendon allows the subdeltoid bursa to directly approximate with the surface of the humeral head. The subdeltoid bursa may be thickened (up to 5 mm), and a bursal effusion is seen lateral to the greater tuberosity. Associated joint effusion can be seen along the biceps tendon.

In small full-thickness tears, a localized absence of the rotator cuff is noted. Tears are commonly seen 1 cm lateral to the biceps tendon. When there is minimal or no retraction of the torn edges of a rotator cuff tear, or when thickened bursa fills the tear, there may be no sonographically visible defect. In such cases, pressure can be applied with the transducer to separate the edges and thereby create a visible concavity in the deep surface of the deltoid (Figure 4). The normal rotator cuff should be entirely noncompressible.

- Partial-thickness tears usually appear as a small hypoechoic defect with an associated hyperechoic focus. The hyperechoic component of partial tears likely is caused by additional interfaces created by the retraction of torn fibers. Pitting and irregularity of the bony surface of the greater tuberosity is a common associated finding. Unlike small full-thickness tears, most partial-thickness tears will not demonstrate any changes when compression is applied with the transducer.

- Abnormality of cuff echogenicity can be diffuse or focal. Diffuse abnormality is seen in inflammation or fibrosis, and this finding is not very reliable. A major disparity of cuff echogenicity is seen with cuff attrition associated with partial-thickness tear. Focal abnormality of cuff echogenicity is associated with small full- and partial-thickness tears. The area of increased echogenicity results from granulation tissue, hypertrophied synovium, and hemorrhage. Either intratendinous hypoechoic focus or a dominant echogenic focus is suggestive of partial-thickness tears.

Secondary ultrasound findings in rotator cuff tears include the following:

- Effusion: 70% of patients with an isolated subdeltoid bursal effusion and about 60% of patients with isolated biceps tendon sheath effusions will have an associated rotator cuff tear,2 as will 95% of patients who have fluid in the biceps tendon sheath and the subdeltoid bursa.6

- Cartilage interface sign: The articular hyaline cartilage of the humeral epiphysis appears as the high-level echo interface. These high-level echoes occur when the diseased tendon substance allows unimpeded sound insonance, which highlights the cartilage interface (Figure 5).

- Greater tuberosity irregularity is another indicator of rotator cuff tear. Ultrasound showed greater tuberosity to be irregular in 90% of shoulders with tears and irregular in only 11% of shoulders without a tear. When greater tuberosity was irregular, ultrasound showed 75% of shoulders to have rotator cuff tears; when the tuberosity was normal, 96% of rotator cuffs were normal on ultrasound.7

OTHER FINDINGS

Subscapularis tendon tears: an isolated subscapularis tendon tear is rare and is often caused by trauma when a person is moving and falls with the shoulder in extension and abduction. Particularly in children, the tear often includes an avulsion fragment of the lesser tuberosity, on which the subscapularis inserts. The accuracy of ultrasound for diagnosing subscapularis tears is the same as for supraspinatus tears.8

Biceps tendon pathology: Biceps tendon effusion appears as an anechoic or hypoechoic halo completely or partially surrounding the extra-articular portion of the biceps sheath. Rupture of the biceps tendon manifests sonographically as nonvisualization of the tendon at the level of the tendon groove and within the joint space. In some cases, the torn tendon will retract down the arm and the attached muscle will be shortened and distorted. In other cases, portions of the tendon will still be seen in or immediately below the groove, especially on longitudinal view. In dislocation of the biceps tendon, the tendon migrates medial to the lesser tuberosity, and on ultrasound the tendon groove is empty. This can sometimes be associated with a tear of the subscapularis tendon. The subluxation of the biceps tendon so that it is partially out of the groove can also occur. Inflammation of the biceps tendon can manifest as fluid in the tendon sheath, tendon enlargement, and hypervascularity on color Doppler.

Other conditions: With calcific tendinitis, tendon appears as a hyperechoic focus that attenuates the sound and usually produces shadowing. Suprascapular ganglion cyst may compress the suprascapular nerve and produce pain. It appears as cystic lesion in the superior aspect of the shoulder between the supraspinatus muscle and the glenoid. Fracture of the greater tuberosity appears as a cortical step off or defect at the junction of the tuberosity and the humeral shaft.

When ultrasound is performed on a symptomatic rotator cuff tear, a contralateral asymptomatic rotator cuff tear is occasionally seen. Ultrasound follow-up over a five-year period to assess the risk of development of symptoms and tear progression showed that 51% of the previously asymptomatic patients became symptomatic over a mean period of 2.8 years.9

POSTSURGICAL EVALUATION

Ultrasound can be a valuable tool in the postsurgical evaluation of the shoulder. The area of tendon repair appears as a heterogeneous fenestrated segment. The trough or tunnel where the postsurgical tendon is implanted on the greater tuberosity can often be seen. Cuff thinning and abnormalities in shape do not return to normal (pretear condition) after surgery.

Ultrasound can help assess integrity of the tendon or identify new tears. Recurrent tears of the cuff are usually full-thickness tears. Sensitivity of 100% and specificity of 94% have been reported with ultrasound in detecting retears.10 In another study, ultrasound detected postsurgical supraspinatus retears in 16% of shoulders. The thinning of the rotator cuff was seen in 23.5% of shoulders and was classified as pathologic if a difference existed between the diameter of the rotator cuff measured in sixth to eighth postoperative week and the diameter measured at the time of follow-up.

Ultrasound offers critical evaluation of both surgical results and postoperative physical therapy protocol.11 For frozen shoulder, mobilization under anesthesia has been used as an effective therapy, although some have voiced concern about rotator cuff tears. Ultrasound has been used in monitoring the integrity of the rotator cuff in these patients, and tears were noted in about 5% of patients.12

Management of the shoulder depends on the presence of cuff injury and the size of the tear. Treatment options include conservative nonsurgical management for patients with intact or partially torn cuff, arthroscopic decompression of the coracoacromial space for those not responding to nonsurgical management, and a range of surgical techniques to repair full-thickness tears. Ultrasound has shown a high correlation between sonographic classification of rotator cuff injury and surgical findings. The selection of appropriate programs can thus reliably be based on the sonographic classification.13

MRI AND ARTHROGRAPHY

In screening patients with chronic shoulder pain, surgery confirmed ultrasound findings of rotator cuff tears in 90.9% (100% sensitivity) when the ultrasound criteria of cuff disappearance and convexity loss were used. This resulted in 9.1% false positives, however, when ultrasound criteria of hyperechoic lesions and thinning were used. Hence, in the latter cases, potential surgical patients must be submitted to MRI.14 Along with arthrography, MRI can assess lesions of the labrum (Bankart's lesions, SLAP lesions) and the joint capsule, as well as muscle atrophy. In these cases, MRI could be the imaging modality of choice.15

Prospective study comparing ultrasound with arthrography has shown that ultrasound can be a noninvasive, painless, and cost-effective alternative. Ultrasound achieved 90% sensitivity and 91% specificity, with positive accuracy of 87% and negative accuracy of 93%. Although it is difficult on ultrasound to differentiate between tears and degenerative changes, an experienced sonologist can often make that differentiation.16 One of the most frequent causes of shoulder pain is noncalcific tendinitis secondary to impingement. Clinical examination is often nonspecific and arthrography is normal. Ultrasound can demonstrate abnormalities within the intact rotator cuff tendon, including changes in echogenicity and thickness of the tendon. The patterns of abnormality demonstrated correlate with pathologic changes seen in tendinitis.17

In experienced hands, ultrasound is as sensitive as arthrography and MRI for detecting rotator cuff tears and abnormalities of the biceps tendon. Ultrasound should be used as the initial imaging test when the primary question is one of these conditions.

Ultrasound may also someday replace some of the diagnostic arthroscopic studies performed. As a recent study comparing ultrasound with arthroscopy showed, ultrasound had a sensitivity of 100% and a specificity of 85%, with an overall accuracy of 96%.3

Ultrasound can be performed faster than MR and is more cost-effective, and its excellent demonstration of the tendon structure should make it the first-line modality for evaluating tendons. A focal dynamic exam of the symptomatic area with clinical correlation helps pinpoint the diagnosis. Although ultrasound is highly user dependent, its applications in shoulder and other musculoskeletal imaging will definitely increase as more experience is gained. We have found that getting the referring physician involved in the ultrasound study has helped increase the awareness of the availability of this diagnostic modality.

References

1. Wiener SN, Seitz WH. Sonography of the shoulder in patients with tears of the rotator cuff: Accuracy and value for selecting
surgical options. AJR 1993;160:103-107.
2. Van Holsbeeck MT, Kolowich PA, Eyler WR, et al. US depiction of partial-thickness tear of the rotator cuff. Radiology 1995;197:443-446.
3. Teefey SA, Hasan SA, Middleton WD, et al. Ultrasonography of the rotator cuff. A comparison of ultrasonographic and arthroscopic findings in one hundred consecutive cases. J Bone Joint Surg Am 2000;82:498-504.
4. Gohlke F. Ultrasonographic appearance of the rotator cuff in elderly subjects. Orthopade 1993;22(5):288-293.
5. Arcuni SE. Rotator cuff pathology and subacromial impingement. Nurse Pract 2000;25(5):58,61,65-66 passim.
6. Hollister MS, Mack LA, Patten RM, et al. Association of sonographically detected subacromial/subdeltoid bursa effusion and intraarticular fluid with rotator cuff tear. AJR 1995; 165;605-608.
7. Wohlwend JR, van Holsbeeck M, Craig J, et al. The association between irregular greater tuberosities and rotator cuff tears: a sonographic study. AJR 1998;171(1):229-233.
8. Farin PU, Jaroma H. Sonographic detection of tears of the anterior portion of the rotator cuff (subscapularis tendon tears). J Ultrasound Med 1996;16:221-225.
9. Yamaguchi K, Tetro AM, Blam O, et al. Natural history of asymptomatic rotator cuff tears: A longitudinal analysis of asymptomatic tears detected sonographically. J Shoulder Elbow Surg 2001;10(3):199-203.
10. Masaoka S, Hashizume H, Senda M, et al. Ultrasonographic analysis of shoulder rotator cuff tears. Acta Med Okayama 1999;53(2): 81-89.
11. Fabis J. Ultrasonographic evaluation of the rotator cuff after its surgical reconstruction. Chir Narzadow Ruchu Ortop Pol 1999;64(5);527-531.
12. Weber M, Prim J, Bugglin R, et al. Long-term follow up to patients with frozen shoulder after mobilization under anesthesia, with special reference to the rotator cuff. Clin Rheumatol 1995;14(6):686-691.
13. Wiener SN, Seitz WH Jr. Sonography of the shoulder in patients with tears of the rotator cuff: accuracy and value for selecting surgical options. AJR 1993;160(1):103-107; discussion 109-110.
14. Chiodi E, Morini G. Chronic shoulder pain. Ultrasonography versus surgery. Radiol Med (Torino) 1995;89(5):600-603.
15. Zanetti M, Hodler J. Imaging of degenerative and posttraumatic disease in the shoulder joint with ultrasound. Eur J Radiol 2000;35(2):119-125.
16. Emamian SA, Jensen DB. Ultrasonography of the rotator cuff of the shoulder. Ugeskr Laeger 1992;154(12):766-770.
17. Crass JR, Craig EV, Feinberg SB. Clinical significance of sonographic findings in the abnormal but intact rotator cuff: a preliminary report. J Clin Ultrasound 1988;16(9):625-634.
DR. RAWOOL is a research associate in radiology at Thomas Jefferson University Hospital in Philadelphia.
Dr. Rawool has no significant financial arrangement or affiliation with any manufacturer of any pharmaceutical or medical device and is not affiliated in any manner with any provider of any commercial medical or healthcare profession service.
Jointly sponsored by CME, Inc. and CMP Healthcare Group
Made possible by an unrestricted educational grant from Toshiba America Medical Systems
---

LEARNING OBJECTIVES
Upon completion of this activity, participants should be able to:
- Describe standard ultrasound technique for performing shoulder ultrasound.
- Identify normal shoulder ultrasound anatomy.
- Recognize various rotator cuff pathologies, including full and partial thickness tears, inflammation, and fluid collections.
- Summarize current trends in shoulder ultrasound imaging.

---

To complete post-test activity and evaluation and register for earned
category 1 CME credit at no charge to you, go to diagnosticimaging.com. If you would like to receive the evaluation form and post-test by mail or fax,
please call toll free at 800/778-2142.
This activity has been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education through the joint sponsorship of CME, Inc. and CMP Healthcare Group. CME, Inc. is accredited by the ACCME to provide continuing medical education for physicians.
CME, Inc. designates this educational activity for a maximum of 1.0 hours in category 1 credit toward the AMA Physician's Recognition Award. Each physician should claim only those hours of credit that he/she actually spent on the educational activity.
Activities that have been approved for AMA category 1 credit and are relevant to the radiologic sciences will be awarded category A credit on a one for one basis by the American Society of Radiologic Technologies (ASRT).