Chronic Shoulder Pain

Procedure Appropriateness Category Relative Radiation Level Radiography shoulder Usually Appropriate ☢ US shoulder May Be Appropriate O Image-guided anesthetic +/corticosteroid injection shoulder or surrounding structures Usually Not Appropriate Varies MR arthrography shoulder Usually Not Appropriate O MRI shoulder without and with IV contrast Usually Not Appropriate O MRI shoulder without IV contrast Usually Not Appropriate O Bone scan shoulder Usually Not Appropriate ☢☢☢ CT shoulder with IV contrast Usually Not Appropriate ☢☢☢ CT shoulder without and with IV contrast Usually Not Appropriate ☢☢☢ CT shoulder without IV contrast Usually Not Appropriate ☢☢☢ CT arthrography shoulder Usually Not Appropriate ☢☢☢☢ FDG-PET/CT skull base to mid-thigh Usually Not Appropriate ☢☢☢☢


Summary of Literature Review
Introduction/Background Chronic shoulder pain lasting >6 months is a common presenting complaint, particularly in patients ≥40 years of age [1].Along with a thorough history and physical examination, imaging plays a critical role in determining the source of chronic shoulder pain.Common pain generators in the shoulder include the rotator cuff tendons, biceps tendon, labrum, glenohumeral articular cartilage, acromioclavicular joint, bones, suprascapular and axillary nerves, and the joint capsule/synovium.Imaging can help to guide appropriate therapy, which can range from conservative measures such as physical therapy and corticosteroid injections to more invasive procedures such as arthroscopy or surgery.This document focuses on the imaging approach to patients with suspected chronic mechanical shoulder pain.See the separate ACR Appropriateness Criteria ® topics on "Chronic Extremity Joint Pain-Suspected Inflammatory Arthritis" [2], "Imaging After Shoulder Arthroplasty" [3], "Primary Bone Tumors" [4], "Soft-Tissue Masses" [5], and "Shoulder Pain-Traumatic" [6] if further guidance is needed for potential alternative causes of shoulder pain.

Special Imaging Considerations
Several studies have looked for ways to shorten typical shoulder conventional MRI or MR arthrogram protocols while maintaining diagnostic accuracy.Isotropic 3-D imaging has been increasingly used as a substitute for conventional multiplanar 2-D MRI, with a similar sensitivity and specificity for rotator cuff tear and labral abnormalities at a lower scan time [7,8].Faster scans can also be performed using 2-D fast spin echo sequences with parallel imaging, which produce similar findings as seen on conventional MRI [9].

Initial Imaging Definition
Initial imaging is defined as imaging at the beginning of the care episode for the medical condition defined by the variant.More than one procedure can be considered usually appropriate in the initial imaging evaluation when: • There are procedures that are equivalent alternatives (ie, only one procedure will be ordered to provide the clinical information to effectively manage the patient's care)

OR
• There are complementary procedures (ie, more than one procedure is ordered as a set or simultaneously where each procedure provides unique clinical information to effectively manage the patient's care).The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels.Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization.
Reprint requests to: publications@acr.orgChronic Shoulder Pain Discussion of Procedures by Variant Variant 1: Chronic shoulder pain.Initial imaging.

Bone Scan Shoulder
There is insufficient evidence to support the use of bone scan shoulder in the initial evaluation of chronic shoulder pain.

CT Arthrography Shoulder
There is insufficient evidence to support the use of CT arthrography shoulder in the initial evaluation of chronic shoulder pain.

CT Shoulder With IV Contrast
There is insufficient evidence to support the use of CT shoulder with intravenous (IV) contrast in the initial evaluation of chronic shoulder pain.

CT Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of CT shoulder without and with IV contrast in the initial evaluation of chronic shoulder pain.

CT Shoulder Without IV Contrast
There is insufficient evidence to support the use of CT shoulder without IV contrast in the initial evaluation of chronic shoulder pain.

FDG-PET/CT Skull Base to Mid-Thigh
There is insufficient evidence to support the use of fluorine-18-2-fluoro-2-deoxy-D-glucose (FDG)-PET/CT skull base to mid-thigh in the initial evaluation of chronic shoulder pain.

Image-Guided Anesthetic +/-Corticosteroid Injection Shoulder or Surrounding Structures
There is insufficient evidence to support the use of image-guided anesthetic +/-corticosteroid injection of the shoulder in the initial evaluation of chronic shoulder pain.

MR Arthrography Shoulder
There is insufficient evidence to support the use of MR arthrography shoulder in the initial evaluation of chronic shoulder pain.

MRI Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of MRI shoulder without and with IV contrast in the initial evaluation of chronic shoulder pain.

MRI Shoulder Without IV Contrast
There is insufficient evidence to support the use of MRI shoulder without IV contrast in the initial evaluation of chronic shoulder pain.

Radiography Shoulder
Most literature suggests that radiographs should be the initial imaging study in patients with shoulder pain [1,10].
Standard radiographic examination consists of a combination of different projections typically including anteroposterior (AP) internal rotation, AP external rotation, Grashey (oblique AP in the plane of the glenohumeral joint), scapular "Y," and axillary views [10].Radiographs can evaluate for unexpected fracture, unsuspected aggressive bony lesion, osteoarthritis, inflammatory arthropathy, osteonecrosis, or calcium hydroxyapatite deposition.Accurate interpretation of radiographs can often obviate additional imaging or be complementary to subsequent imaging studies.A high-riding humeral head on radiographs is highly associated with chronic rotator cuff tear [11], particularly if seen in combination with sclerosis of the inferior acromion, greater tuberosity cysts, a lateral acromial spur, and cortical irregularity or sclerosis of the greater tuberosity [12][13][14].

US Shoulder
The literature suggests that radiographic evaluation should generally be the first imaging study ordered in patients with shoulder pain [1,[10][11][12][13][14].Some practitioners may use ultrasound (US) as an initial imaging study when rotator cuff disease/impingement is suggested by the clinical examination.Chronic Shoulder Pain Variant 2: Chronic shoulder pain.Suspect rotator cuff disorders or subacromial subdeltoid bursitis (no prior surgery).Initial radiographs normal or inconclusive.Next imaging study.

Bone Scan Shoulder
There is insufficient evidence to support the use of bone scan shoulder in the evaluation of chronic shoulder pain when rotator cuff disorder or subacromial subdeltoid bursitis is suspected.

CT Arthrography Shoulder
CT arthrography has shown good diagnostic performance for depicting rotator cuff tendon tears with a reported respective sensitivity and specificity of 99% and 100% for the supraspinatus tendon, 97.4% and 99.5% for the infraspinatus tendon, and 64.7% and 98.2% for the subscapularis tendon [15].CT arthrography provides a comparable alternative to MR arthrography in the evaluation of rotator cuff tendon tears [16,17].CT arthrography is able to detect very subtle articular surface cuff defects, but, because of lack of contrast extension from the glenohumeral joint space, it is less effective at demonstrating bursal surface or intrasubstance tears, which are typically well seen with MRI or US [18].

CT Shoulder With IV Contrast
There is insufficient evidence to support the use of CT shoulder with IV contrast in the evaluation of chronic shoulder pain when rotator cuff disorder or subacromial subdeltoid bursitis is suspected.

CT Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of CT shoulder without and with IV contrast in the evaluation of chronic shoulder pain when rotator cuff disorder or subacromial subdeltoid bursitis is suspected.

CT Shoulder Without IV Contrast
CT shoulder without IV contrast is of limited utility when evaluating for rotator cuff tear or subacromial subdeltoid bursitis.Noncontrast CT can sometimes demonstrate fatty infiltration of rotator cuff muscles with a corresponding rotator cuff tear, although the overall sensitivity of this finding is low [19].CT may demonstrate a high-riding humeral head or remodeling of the acromial undersurface in patients with large rotator cuff tears.However, this finding is generally seen on initial radiographs, and CT does not allow for characterization of the underlying tear because CT lacks the soft-tissue contrast resolution necessary to adequately evaluate the rotator cuff tear itself.CT may be able to directly demonstrate some large rotator cuff tears as well as fatty infiltration of the rotator cuff musculature; however, it was found to have a sensitivity of 20% for full thickness rotator cuff tears in one small study [19].Large amounts of fluid in the subacromial subdeltoid bursa can sometimes be visualized with CT.

FDG-PET/CT Skull Base to Mid-Thigh
There is insufficient evidence to support the use of FDG-PET/CT skull base to mid-thigh in the evaluation of chronic shoulder pain when rotator cuff disorder or subacromial subdeltoid bursitis is suspected.

Image-Guided Anesthetic +/-Corticosteroid Injection Shoulder or Surrounding Structures
Corticosteroid injection is sometimes used before advanced imaging is obtained but may serve as a better diagnostic tool in patients with subacromial subdeltoid bursitis as opposed to those patients with suspected rotator cuff tear.US image-guided injection with local anesthetic and corticosteroid into the subacromial-subdeltoid bursa has been shown to result in improvement in pain and range of motion in patients with rotator cuff symptoms [20,21].Subacromial bursal injections are used by some as a diagnostic tool to evaluate for rotator cuff pathology [22], although the diagnostic utility of this approach is unclear in the literature.Glenohumeral joint injection has been found to not have significant diagnostic utility for patients with suspected rotator cuff tear [23].

MR Arthrography Shoulder
MR arthrography provides greater sensitivity and specificity compared to conventional MRI in evaluating partial thickness articular surface tears [24][25][26].Partial bursal and purely intrasubstance tears should have similar rates of detection on conventional MRI and MR arthrography because the diagnosis is dependent on the same fluid-sensitive sequences [27].Extension of intra-articular gadolinium into the subacromial subdeltoid bursa can sometimes differentiate a high-grade partial tear from a full thickness tear [28].However, the advantages of MR arthrography over conventional MRI for detecting rotator cuff tear must be weighed against the invasive nature of the arthrogram procedure [29].Arthrograms with saline have been shown in some studies to have similar accuracy for rotator cuff tear as those performed with dilute gadolinium [30,31].Some authors suggest that adding an abduction external rotation imaging position to an MR arthrogram improves the accuracy in diagnosing partial articular surface tears [32,33], although this positioning is often omitted because of patient discomfort.Chronic Shoulder Pain

MRI Shoulder Without and With IV Contrast
IV injection of gadolinium can be used to enhance the highly vascular synovium producing an indirect arthrogram effect [34].Indirect MR arthrography has been shown to have a slightly higher specificity for diagnosing articularsurface partial thickness tears of the supraspinatus and infraspinatus tendons, as well as a greater specificity for diagnosing subscapularis tendon tears [35], relative to noncontrast MRI, and has shown high accuracy relative to arthroscopy [36].Although the accuracy of indirect MR arthrography has been described in the literature, it is not commonly used in clinical practice.

MRI Shoulder Without IV Contrast
MRI of the shoulder is a highly accurate tool in the assessment of the rotator cuff.MRI findings of tendinopathy consist of intermediate increased signal and tendon thickening [28].Rotator cuff tears, particularly full thickness tears, can be reliably identified using conventional MRI with high sensitivity and specificity [37][38][39].Sensitivity and specificity for detecting partial thickness bursal surface tears is similar for conventional MRI and MR arthrography [27].Although MR arthrography may have a slightly higher sensitivity and specificity than conventional MRI for diagnosing rotator cuff tears, particularly partial thickness articular surface tears, this difference in accuracy must be weighed against the need for an invasive arthrogram procedure [29].Tendon retraction, muscle atrophy, and fatty infiltration are associated important findings that are well seen on MRI and predict reparability of rotator cuff tear [40].Axial images can be helpful in determining rotator cuff tear shape [41], which can aid in the selection of a particular surgical technique.In addition to evaluating the rotator cuff itself, MRI can detect extra-articular abnormalities that may predispose to shoulder impingement such as acromioclavicular degenerative change or a subacromial spur.Fluid accumulation within the subacromial subdeltoid bursa is visualized on MRI and is often a nonspecific finding seen with a rotator cuff tear or an underlying inflammatory disorder [42].

Radiography Shoulder Additional Views
The rotator cuff tendons cannot be directly assessed radiographically, regardless of the view.Secondary findings of rotator cuff tear such as a high-riding humeral head or changes at the greater tuberosity can be seen on standard radiographic views.There have been several reports assessing special views for the evaluation of potential impingement from the anterior acromion.Outlet views can be used to determine acromial shape and have been shown to be more accurate than a single-slice MRI [43].The Rockwood and cassette tilt views are angled frontal projections that can be used to detect anterior acromial osteophytes [44].The subacromial subdeltoid bursa is a softtissue potential space and is not directly visualized on radiographs.There is a peribursal fat plane between the rotator cuff tendons and the deltoid muscle; however, obliteration of the peribursal fat stripe is not a specific indicator of shoulder pathology and may be seen in normal subjects [42].

US Shoulder
US has been found to have a high accuracy in assessing rotator cuff disease [38,45], including the subscapularis tendon [46].Meta-analyses have found a sensitivity of US ranging from 85% to 95% and a specificity ranging from 72% to 92%, which is similar to noncontrast MRI and slightly lower than MR arthrography [24,47].Fatty infiltration within the rotator cuff muscles can be adequately demonstrated with US [48].US may; however, be less accurate than MRI in measuring tear size and the degree of retraction in larger tears [49].US allows excellent evaluation of the bursae situated about the shoulder, particularly the subacromial subdeltoid bursa [50].Color flow imaging can demonstrate hyperemia in cases of bursitis with a significant inflammatory component [51].

Variant 3: Chronic shoulder pain. Radiographs demonstrate calcific tendinopathy or calcific bursitis. Next imaging study. Bone Scan Shoulder
There is insufficient evidence to support the use of bone scan shoulder in the evaluation of chronic shoulder pain when radiographs demonstrate calcific tendinosis or calcific bursitis.Cases with intraosseous extension of calcium hydroxyapatite can potentially demonstrate increased radionuclide uptake due to bony inflammatory change, which can potentially be confused with a neoplastic process and result in diagnostic confusion.

CT Arthrography Shoulder
There is insufficient evidence to support the use of CT arthrography shoulder in the evaluation of chronic shoulder pain when radiographs demonstrate calcific tendinopathy or calcific bursitis.Hyperdense intratendinous calcium may resemble intratendinous contrast from a partial thickness articular-sided rotator cuff tear, which could potentially result in diagnostic confusion.The role of CT arthrography is limited to cases in which there is concern Chronic Shoulder Pain for a concomitant rotator cuff tear and, even in this situation, may be lacking in utility in cases of partial thickness interstitial or bursal surface tear [18].

CT Shoulder With IV Contrast
There is insufficient evidence to support the use of CT shoulder with IV contrast in the evaluation of chronic shoulder pain when radiographs demonstrate calcific tendinopathy or calcific bursitis.

CT Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of CT shoulder without and with contrast in the evaluation of chronic shoulder pain when radiographs demonstrate calcific tendinopathy or calcific bursitis.

CT Shoulder Without IV Contrast
Calcific hydroxyapatite deposition in the rotator cuff tendons is easily visualized by CT; however, the diagnosis of calcium hydroxyapatite is typically able to be made on radiographs alone.CT can be helpful in evaluating osseous involvement [52] and is the most accurate modality in evaluating the consistency of the deposit, which may be helpful when planning intervention [53] but is not routinely obtained.

FDG-PET/CT Skull Base to Mid-Thigh
There is insufficient evidence to support the use of FDG-PET/CT skull base to mid-thigh in the evaluation of chronic shoulder pain when radiographs demonstrate calcific tendinopathy or calcific bursitis.Areas of calcium hydroxyapatite deposition in tendon, bursa, or bone can result in increased FDG uptake, which can potentially be confused with a neoplastic process and result in diagnostic confusion.

Image-Guided Anesthetic +/-Corticosteroid Injection Shoulder or Surrounding Structures
If calcium hydroxyapatite is seen on radiographs, calcium hydroxyapatite lavage with imaging guidance can be performed both for both therapeutic and diagnostic purposes.US or, less commonly, fluoroscopy can be used for guidance.US-guided calcium lavage has been shown to be a highly effective percutaneous treatment in patients with suspected calcific tendinitis [54][55][56][57][58][59][60][61][62].Outcomes are similar whether the procedure is performed with the 1 or 2 needle technique [63].Injection of corticosteroid into the subacromial subdeltoid bursa as part of the barbotage procedure has been shown to have significant impact on pain and function 3 months after the procedure [64,65].The degree of calcium removal does not necessarily correlate with pain improvement from the procedure [66].Lavage may be less effective at providing pain relief for patients with intraosseous migration of calcium hydroxyapatite [67].Given the high efficacy of calcium hydroxyapatite lavage in the rotator cuff, if patient symptoms do not resolve following calcium hydroxyapatite lavage, then other diagnostic entities for shoulder pain should be considered and further imaging evaluation can be performed subsequently.

MR Arthrography Shoulder
Calcium hydroxyapatite may be difficult to visualize on MRI given the lack of contrast between hypointense calcium and an adjacent normal tendon.However, MR arthrography can be used to identify a concomitant rotator cuff tear.In the setting of calcific tendonitis, 93% of rotator cuff tears are partial thickness [68].Given the improved ability of MR arthrography to detect partial thickness articular-surface tears of the rotator cuff compared to conventional shoulder MRI [24][25][26], there may be an advantage to performing MR arthrography in patients with calcific tendinopathy [69].

MRI Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of MRI shoulder without and with IV contrast in the evaluation of chronic shoulder pain when radiographs demonstrate calcific tendinopathy or calcific bursitis.

MRI Shoulder Without IV Contrast
The signal void of calcium hydroxyapatite deposition can be difficult to distinguish from the hypointense signal in normal rotator cuff tendon, thus making identification of calcific tendonitis by MRI alone difficult in some cases.However, MRI can be used in calcific tendonitis cases to evaluate the extent of adjacent soft-tissue and bone abnormalities as well as exclude other causes of shoulder pain.Patients with calcific tendonitis on radiographs have up to a 56% incidence of concomitant rotator cuff tear seen on MRI [68].Intraosseous migration of calcium hydroxyapatite can produce extensive marrow edema, which can be confused with a neoplastic process [52,70].

Radiography Shoulder Additional Views
Calcium hydroxyapatite deposition in tendon or bursa is typically well seen on conventional radiographic projections, without a need for specialized additional views.

US Shoulder
US is reported to be 98% sensitive and 94% specific for calcium hydroxyapatite in the rotator cuff tendons [71], although identification may be challenging if the area of calcium deposition is small.At times US may have difficulty distinguishing calcific deposits from heterotopic ossification because both can result in dense shadowing at US. US can be used if there is concern for concomitant rotator cuff tendon tear and may be a better modality than MRI in this situation given the difficulty of distinguishing calcium from normal tendon at MRI.

Variant 4: Chronic shoulder pain. Suspect labral pathology or shoulder instability. Initial radiographs normal or inconclusive. Next imaging study. Bone Scan Shoulder
There is insufficient evidence to support the use of bone scan shoulder in the evaluation of chronic shoulder pain when labral pathology or shoulder instability is suspected.

CT Arthrography Shoulder
CT arthrograms have been shown to be an accurate modality in the assessment of shoulder instability because of its depiction of osseous, cartilaginous, and labroligamentous injuries [18,72].Labral evaluation in the postoperative setting may be better with CT arthrography than MRI in patients with metallic suture anchors [73] and has been used to evaluate for healing after superior labrum anterior-to-posterior (SLAP) repair [74].Dual-energy CT arthrography has a similar diagnostic performance as MR arthrography [75].Although CT arthrography aids in the evaluation of underlying soft tissues, the injected contrast may limit evaluation of the underlying osseous structures, and cortical fragments in the injected contrast solution often have a similar attenuation to cortical bone [76].Other limitations include the need for an invasive arthrography procedure and poor assessment of bone marrow edema [72].

CT Shoulder With IV Contrast
There is insufficient evidence to support the use of CT shoulder with IV contrast in the evaluation of chronic shoulder pain when labral pathology or shoulder instability is suspected.

CT Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of CT shoulder without and with IV contrast in the evaluation of chronic shoulder pain when labral pathology or shoulder instability is suspected.

CT Shoulder Without IV Contrast
CT is helpful for assessing bony glenoid deficiency as well as the size of a Hill-Sachs deformity when surgery is a consideration but is not effective at evaluating the labrum.CT is often preferable to MRI when evaluating for small fracture fragments of the glenoid rim and assessing bone stock in patients with recurrent dislocation [77][78][79].Threedimensional reconstructions can be used to quantify osseous defect width and glenoid surface area [80,81].CT can also be used to measure the size of a Hill-Sachs defect [82,83], which has become more important with the development of the glenoid track concept for determining instability [84,85].In the postoperative setting, CT can be used to evaluate integration of bone graft from a coracoid transfer with the underlying glenoid [86,87].

FDG-PET/CT Skull Base to Mid-Thigh
There is insufficient evidence to support the use of FDG-PET/CT skull base to mid-thigh in the evaluation of chronic shoulder pain when labral pathology or shoulder instability is suspected.

Image-Guided Anesthetic +/-Corticosteroid Injection Shoulder or Surrounding Structures
Patients with labral tears have been found to have no significant decrease in pain with glenohumeral injection relative to other patients, and, in fact, patients with labral tears may have less pain relief than those with an intact labrum [23].Thus, corticosteroid injection does not have a significant diagnostic benefit in distinguishing patients with labral tears from other pathology.

MR Arthrography Shoulder
Multiple meta-analyses have found MR arthrography to have a higher sensitivity (80%-87% versus 63%-76%) and specificity (91%-92% versus 87%) compared to conventional MRI for the detection of SLAP tears [88,89].A 2018 meta-analysis found arthrography to be slightly more sensitive for anterior labral lesions (87% compared to 83%) [90].Provocative positioning maneuvers, including abduction external rotation, can result in improved detection of nondisplaced anteroinferior labral tears [91,92], although this positioning may be uncomfortable for many patients.MR arthrography is highly accurate for detection of injury to the inferior glenohumeral ligament, which can be associated with shoulder instability [93].Arthrograms with saline have been shown in some studies to have similar Chronic Shoulder Pain accuracy for a labral tear as those performed with dilute gadolinium [30,31].Although diagnosing recurrent labral tears after prior labral repair is challenging, MR arthrography is particularly helpful in this scenario in which discrete fluid or gadolinium signal within or underlying the labrum, absent labrum, and paralabral cyst are all highly suggestive of recurrent tear [94], with a greater sensitivity than seen with conventional MRI [73].

MRI Shoulder Without and With IV Contrast
IV injection of gadolinium can be used to enhance the highly vascular synovium producing an indirect arthrogram effect [34].Indirect arthrography has been found to have a sensitivity of 74% and a specificity of 67% for SLAP tear, which is lower than for direct MR arthrography [89].Although the accuracy of indirect MR arthrography has been described in the literature, it is not commonly used in clinical practice.

MRI Shoulder Without IV Contrast
Conventional MRI with 3T imaging is highly accurate for the detection of a labral tear, with a sensitivity of 83% and a specificity of 99% to 100% for a labral tear [90].MR arthrography adds additional benefit to conventional shoulder MRI because it may identify some labral tears that are not seen with conventional MRI [95].The increased accuracy with MR arthrography should be balanced against the need for an invasive arthrogram procedure.
Although evaluation of glenoid bony integrity is typically better with CT imaging, some practitioners have used ultrashort or zero echo time sequences [96][97][98], 3-D sequences [99], or conventional MRI sequences [100] to measure glenoid bone loss with a high concordance to CT images.

Radiography Shoulder Additional Views
There are several options for additional radiographic views in patients with suspected instability [1].The West Point and Garth views can demonstrate the anterior inferior glenoid at a better advantage in patients with a suspected Bankart fracture.A Stryker notch view can sometimes demonstrate a Hill-Sachs lesion that is difficult to see on conventional projections.Although these additional projections are used by some practitioners, they do not usually alleviate the need for advanced imaging.

US Shoulder
The posterior labrum can be reliably visualized with US, with posterior labral tears potentially visualized particularly if the joint is distended [101].However, US is of limited benefit in assessing the anterior labroligamentous complex, superior labrum, articular surface, and subchondral bone.

Next imaging study. Bone Scan Shoulder
There is insufficient evidence to support the use of bone scan shoulder in the evaluation of chronic shoulder pain when adhesive capsulitis is suspected.

CT Arthrography Shoulder
There is insufficient evidence to support the use of CT arthrography shoulder in the evaluation of chronic shoulder pain when adhesive capsulitis is suspected.

CT Shoulder With IV Contrast
There is insufficient evidence to support the use of CT shoulder with IV contrast in the evaluation of chronic shoulder pain when adhesive capsulitis is suspected.

CT Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of CT shoulder without and with IV contrast in the evaluation of chronic shoulder pain when adhesive capsulitis is suspected.

CT Shoulder Without IV Contrast
There is insufficient evidence to support the use of CT shoulder without IV contrast in the evaluation of chronic shoulder pain when adhesive capsulitis is suspected.

FDG-PET/CT Skull Base to Mid-Thigh
Increased PET uptake in the rotator interval and inferior joint capsule is moderately associated with adhesive capsulitis [102][103][104].However, this is generally an incidental finding on PET examinations performed for other reasons, and PET/CT is not typically performed specifically for this indication.

Image-Guided Anesthetic +/-Corticosteroid Injection Shoulder or Surrounding Structures
Image-guided injection of the glenohumeral joint for adhesive capsulitis can be performed with either fluoroscopy or US guidance.Image-guided corticosteroid injection has been shown to have a significant reduction in pain and an improvement in the range of motion for patients with adhesive capsulitis [105,106].Use of US guidance for glenohumeral corticosteroid injection may be more accurate and produce a greater improvement in pain and functional scores relative to blind injection [107].Patients with adhesive capsulitis have been found to be the only group with a statistically better outcome after glenohumeral joint injection than other patients; thus a positive response to a glenohumeral injection can be suggestive of the presence of adhesive capsulitis in the appropriate clinical context [23].Fluoroscopic injection can also demonstrate diagnostic findings suggestive of adhesive capsulitis including joint space capacity <10 mL, diminished axillary recess, and irregularity of the capsular margin [108].

MR Arthrography Shoulder
Mengiardi et al [109] reported that thickening of the coracohumeral ligament (specificity 95%, sensitivity 59%), thickening of the joint capsule (specificity 86%, sensitivity 64%), and obliteration of the subcoracoid triangle (specificity 100%, sensitivity 32%) are characteristic MR arthrography findings in adhesive capsulitis.However, fluid within the rotator interval or along the axillary recess due to the arthrogram injection itself can serve as a limitation to detecting findings of adhesive capsulitis on MRI arthrography.

MRI Shoulder Without and With IV Contrast
Contrast-enhanced MRI can increase the conspicuity of axillary recess capsular thickening [110] and rotator interval soft-tissue thickening [111], although there are conflicting data on the relationship of this capsular enhancement with patient symptoms [112].

MRI Shoulder Without IV Contrast
Zhao et al [113] showed that coracohumeral ligament thickening, anterior capsule thickening, and obliteration of the subcoracoid fat are the most characteristic findings of adhesive capsulitis on conventional MRI.Chi et al [114] found a sensitivity of 77% and a specificity of 53% for adhesive capsulitis with only coracohumeral ligament thickening on noncontrast MRI, with a lower sensitivity (23%) but a higher specificity (87%) in patients who had all 3 findings of adhesive capsulitis, including coracohumeral ligament thickening, rotator interval infiltration, and axillary recess thickening.

Radiography Shoulder Additional Views
There is insufficient evidence to support the use of additional radiographic views in the evaluation of chronic shoulder pain when adhesive capsulitis is suspected.

US Shoulder
US in the setting of potential adhesive capsulitis is often used to exclude rotator cuff tear.Although there are several US findings that can be seen in patients with adhesive capsulitis, US is not typically performed specifically for this indication.Thickening of the coracohumeral ligament, increased rotator interval soft tissue, and decreased external rotation on dynamic US are all highly sensitive and specific findings for adhesive capsulitis [115].Thickening of the axillary recess capsule can be accurately measured with US and is associated with adhesive capsulitis [116].Hyperemia with Doppler imaging, particularly with microflow imaging, can also be seen in the subcoracoid fat triangle in patients with adhesive capsulitis [117].One study found that findings of adhesive capsulitis on US were correlated with clinical assessment, whereas findings on MRI were not [108].

Variant 6: Chronic shoulder pain. Suspect biceps tendon abnormality. Initial radiographs normal or inconclusive. Next imaging study. Bone Scan Shoulder
There is insufficient evidence to support the use of bone scan shoulder in the evaluation of chronic shoulder pain when biceps tendon abnormality is suspected.

CT Arthrography Shoulder
Reports on the accuracy of CT arthrography for biceps tendinopathy are variable.Teixeira et al [118] reported a sensitivity of 74% to 79% and a specificity of 93% to 95% for tendinopathy as well as 100% sensitivity and 93% specificity for complete biceps ruptures.However, Rol et al [119] reported a correlation of only 45% to 65% with subsequent arthroscopy in a study consisting mostly of patients who underwent CT arthrography.CT arthrography diagnosis of biceps tendinopathy is primarily based off the change in tendon caliber.Chronic Shoulder Pain

CT Shoulder With IV Contrast
There is insufficient evidence to support the use of CT shoulder with IV contrast in the evaluation of chronic shoulder pain when biceps tendon abnormality is suspected.

CT Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of CT shoulder without and with IV contrast in the evaluation of chronic shoulder pain when biceps tendon abnormality is suspected.

CT Shoulder Without IV Contrast
There is limited literature regarding the use of noncontrast CT to specifically evaluate the biceps tendon.In a small study, pooled sensitivity and specificity for proximal biceps lesions including degeneration, tendon subluxation, and tendon tear using CT were shown to be 31% and 95%, respectively [120].

FDG-PET/CT Skull Base to Mid-Thigh
There is insufficient evidence to support the use of FDG-PET/CT skull base to mid-thigh in the evaluation of chronic shoulder pain when biceps tendon abnormality is suspected.

Image-Guided Anesthetic +/-Corticosteroid Injection Shoulder or Surrounding Structures
Image-guided injection of the biceps tendon sheath using US guidance may be useful in patients with biceps pain for both diagnostic and therapeutic purposes [22].Although either fluoroscopy or US can be used as imaging guidance for biceps tendon injection, US guidance is more accurate [121].US is more accurate in injection location placement compared with blind injections with a lower risk of intratendinous injection [122,123].Many patients with sonographically normal biceps tendons may obtain pain relief with US-guided injection, suggesting that injection response may diagnostically identify patients with biceps pathology that is not seen by US [124].

MR Arthrography Shoulder
MR arthrography may improve the accuracy of MRI in the detection of long head of the bicep tears, although results are mixed.Zanetti et al [125] found MR arthrography to be sensitive and moderately specific in the diagnosis of long head of the biceps tendon disorders with a sensitivity of 92% and a specificity of 56%.However, Loock et al [126] found a lower sensitivity and specificity for MR arthrography detection of biceps abnormalities, with only a 49% sensitivity for tendinopathy.Accuracy for middle-aged patients may be lower [127].MR arthrography is moderately accurate at assessing for biceps tendon subluxation, with a sensitivity of 61% to 87% and a specificity of 83% to 92% [128].

MRI Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of MRI shoulder without and with IV contrast in the evaluation of chronic shoulder pain when biceps tendon abnormality is suspected.

MRI Shoulder Without IV Contrast
Standard noncontrast MRI is limited in detecting bicep abnormalities, including tendinopathy or tearing of the long head of the biceps.Studies have shown MRI to both under-and overdiagnose intra-articular biceps pathology [129][130][131], with sensitivities and specificities of 28% and 84% (partial tear) and 56% and 98% (complete), respectively [130].Using diagnostic criteria of having 2 or more signs of biceps tendinopathy (contour irregularity, diameter change, or abnormal signal) has a sensitivity of 78% and a specificity of 94% [132].It has been reported that normal appearing tendons on MRI frequently have histopathologic findings of tendinopathy [133,134].Biceps tendon subluxation can be diagnosed by displacement of the biceps tendon over the lesser tuberosity [135] and is highly associated with tearing of the subscapularis tendon [136].MRI has a greater specificity for biceps tendon abnormality if there is a coexisting rotator cuff tear [129,137].

Radiography Shoulder Additional Views
There is insufficient evidence to support the use of additional radiographic views in the evaluation of chronic shoulder pain when a biceps tendon abnormality is suspected.

US Shoulder
Apart from the biceps anchor, the intra-articular and extra-articular portions of the biceps tendon are well seen on routine US imaging.Tendinopathy, tears and rupture, tenosynovitis, subluxation, and frank dislocation are readily assessed [138,139].Provocative maneuvers may be helpful while observing in real time to assess the degree of subluxation.Chronic Shoulder Pain The presence of glenohumeral osteoarthritis on initial radiographs does not typically require additional imaging unless additional pathology is suspected, such as a rotator cuff tear or surgery is a consideration.

Bone Scan Shoulder
There is insufficient evidence to support the use of bone scan shoulder in the evaluation of chronic shoulder pain when initial radiographs demonstrate osteoarthritis.

CT Arthrography Shoulder
CT arthrography can be helpful in the preoperative planning of total shoulder arthroplasty in patients with severe glenohumeral arthritis.CT arthrography provides excellent imaging of osseous detail allowing for characterization of glenoid morphology, version, and bone loss.Unlike nonarthrogram CT, CT arthrography allows for adequate characterization of a rotator cuff tear [16,17], which may alter the chosen arthroplasty type.Arthrogram contrast also outlines the articular cartilage surface, demonstrating areas of cartilage loss, which cannot be directly visualized with conventional CT.

CT Shoulder With IV Contrast
There is insufficient evidence to support the use of CT shoulder with IV contrast in the evaluation of chronic shoulder pain when initial radiographs demonstrate osteoarthritis.

CT Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of CT shoulder without and with IV contrast in the evaluation of chronic shoulder pain when initial radiographs demonstrate osteoarthritis.

CT Shoulder Without IV Contrast
CT can be helpful in evaluating osseous morphology in cases in which arthroplasty surgery is under consideration for treatment of severe glenohumeral arthritis.CT images are commonly obtained to classify the morphology of the glenoid according to the Walch classification system [140].CT measurements of glenoid version, glenoid bone loss, and humeral head subluxation demonstrate high interrater reliability [141].CT can potentially demonstrate rotator cuff muscle fatty infiltration, which is often associated with rotator cuff tears; however, overall sensitivity for full thickness rotator cuff tears is only 20% for noncontrast CT [19].

FDG-PET/CT Skull Base to Mid-Thigh
There is insufficient evidence to support the use of FDG-PET/CT skull base to mid-thigh in the evaluation of chronic shoulder pain when initial radiographs demonstrate osteoarthritis.

Image-Guided Anesthetic +/-Corticosteroid Injection Shoulder or Surrounding Structures
Intra-articular corticosteroid injection using either fluoroscopy or US guidance is often used in the conservative management of glenohumeral arthritis [122] and may be performed in some situations before advanced imaging is obtained.The diagnostic utility of glenohumeral corticosteroid injection for confirming osteoarthritis as a source of pain is limited, because patients with osteoarthritis have a similar degree of pain improvement (74% rate of clinically relevant improvement at 1 month) as patients with other underlying causes of shoulder pain [23].

MR Arthrography Shoulder
The role of MR arthrography relative to conventional shoulder MRI in the setting of osteoarthritis has not been specifically studied.However, MR arthrography may have use in characterizing the degree of chondral loss, the presence of a full thickness rotator cuff tear, and characterizing glenoid morphology.

MRI Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of MRI shoulder without and with IV contrast in the evaluation of chronic shoulder pain when initial radiographs demonstrate osteoarthritis.

MRI Shoulder Without IV Contrast
MRI can be used to detect the degree of glenohumeral chondral loss, although the diagnosis of arthritis is generally easily made with conventional radiographs [142].MRI can demonstrate the presence of a partial or full thickness rotator cuff tear, which is frequently associated with osteoarthritis.Although CT is most frequently used for evaluating glenoid morphology, glenoid version measured by MRI is generally similar to that measured with CT [143]; however, MRI may be less accurate than CT at identifying the Walch classification of glenoid morphology [144].Chronic Shoulder Pain

Radiography Shoulder Additional Views
Glenohumeral osteoarthritis is typically well visualized on conventional radiographic views, without the need for special views.A standard axillary view can demonstrate posterior subluxation of the humeral head and abnormal glenoid morphology in patients with severe osteoarthritis [145][146][147].A measurement of the critical shoulder angle can be drawn on routine AP radiographs by measuring the angle between a line along the glenoid surface and a line from the inferior glenoid to the lateral acromial margin; patients with a higher measurement may be at high risk of having a full thickness rotator cuff tear [148].

US Shoulder
Although secondary signs of osteoarthritis such as osteophytes, joint effusion, and intra-articular bodies can be seen with US [142], it is not typically performed for this indication.

Bone Scan Shoulder
There is insufficient evidence to support the use of bone scan shoulder in the evaluation of chronic shoulder pain with a history of prior rotator cuff repair when rotator cuff disorder or subacromial subdeltoid bursitis is suspected.

CT Arthrography Shoulder
As in the native shoulder, CT arthrography can be used to evaluate for rotator cuff tear after surgery.In patients with metal in the shoulder, CT arthrography can be beneficial in detecting supraspinatus tears that may be obscured by metal artifact [73].Gadolinium contrast can potentially be drawn up into a mixture with iodinated contrast at the time of the arthrogram procedure, thus allowing for a shift from MRI to CT if artifact on MRI precludes evaluation.

CT Shoulder With IV Contrast
There is insufficient evidence to support the use of CT shoulder with IV contrast in the evaluation of chronic shoulder pain with a history of prior rotator cuff repair when rotator cuff disorder or subacromial subdeltoid bursitis is suspected.

CT Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of CT shoulder without and with IV contrast in the evaluation of chronic shoulder pain with a history of prior rotator cuff repair when rotator cuff disorder or subacromial subdeltoid bursitis is suspected.

CT Shoulder Without IV Contrast
There is insufficient evidence to support the use of CT shoulder without IV contrast in the evaluation of chronic shoulder pain with a history of prior rotator cuff repair when rotator cuff disorder or subacromial subdeltoid bursitis is suspected.

FDG-PET/CT Skull Base to Mid-Thigh
There is insufficient evidence to support the use of FDG-PET/CT skull base to mid-thigh in the evaluation of chronic shoulder pain with a history of prior rotator cuff repair when a rotator cuff disorder or subacromial subdeltoid bursitis is suspected.

Image-Guided Anesthetic +/-Corticosteroid Injection Shoulder or Surrounding Structures
Image-guided corticosteroid injection may be useful for therapeutic purposes in patients with recurrent pain after rotator cuff repair.US-guided injection of corticosteroid into the glenohumeral joint has been found to improve pain and range of motion in patients with prior rotator cuff repair, without increasing the risk of recurrent rotator cuff tear [149].There is insufficient evidence to support the use of image guided injection for diagnostic purposes in evaluation for recurrent cuff tear, but it may be of use for diagnosis in patients with suspected subacromial subdeltoid bursitis.

MR Arthrography Shoulder
MR arthrography is useful in the evaluation of the postoperative rotator cuff.MR arthrography can improve diagnostic accuracy over conventional MRI in patients who have had prior rotator cuff repair due to joint distention and improved signal-to-noise ratio [150].Of note, contrast within the subacromial subdeltoid bursa on routine and MR arthrogram examinations is a nonspecific finding that can potentially be seen in a nonwatertight rotator cuff repair [151].T2 hyperintensity of the repaired rotator cuff tendon is a common postoperative finding, particularly in the first 2 years after surgery, and does not reflect adequacy of tendon healing or clinical outcome [152].

MRI Shoulder Without and With IV Contrast
There is insufficient evidence to support the use of MRI shoulder without and with IV contrast in the evaluation of chronic shoulder pain with a history of prior rotator cuff repair when a rotator cuff disorder or subacromial subdeltoid bursitis is suspected.

MRI Shoulder Without IV Contrast
MRI is useful in the evaluation of the postoperative rotator cuff.The shift from metallic to bioabsorbable suture anchors tends to result in fewer cases with metal artifact implications [153].For patients with metallic suture anchors, there are a variety of commercially available metal artifact reduction sequences, which can limit artifact.MRI can demonstrate complications with suture anchors such as osteolysis or migration [154], even if the suture anchors are nonmetallic.MRI can also demonstrate potential improvements in muscle atrophy and fatty infiltration after rotator cuff repair [155,156].

Radiography Shoulder Additional Views
There is insufficient evidence to support the use of additional radiographic views in the evaluation of chronic shoulder pain with a history of prior rotator cuff repair when rotator cuff disorder or subacromial subdeltoid bursitis is suspected.

US Shoulder
An advantage of US evaluation of the postoperative shoulder is the absence of metal-induced artifact when examining the rotator cuff and adjacent soft tissues.Patients who have undergone rotator cuff repair may show discrete defects on US imaging that may persist for years with some potential for delayed healing [157].Despite these limitations, US still provides an excellent method to assess repair integrity, with a high sensitivity (80.8%) and specificity (100%) for recurrent rotator cuff tear [158].The ability to perform provocative maneuvers can further demonstrate abnormalities that other forms of imaging would be incapable identifying.

Summary of Recommendations
• Variant 1: Radiography is usually appropriate for the initial imaging of chronic shoulder pain.
• Variant 2: US or MR arthrography or MRI without IV contrast is usually appropriate as the next imaging study in the evaluation of chronic shoulder pain when radiographs are normal or inconclusive and rotator cuff disorders or subacromial subdeltoid bursitis (no prior surgery) is suspected.These procedures are equivalent alternatives (ie, only one procedure will be ordered to provide the clinical information to effectively manage the patient's care).
• Variant 3: Image-guided anesthetic +/-corticosteroid injection is usually appropriate as the next imaging study in the evaluation of chronic shoulder pain when radiographs demonstrate calcific tendinopathy or calcific bursitis.
• Variant 4: MR arthrography or MRI without IV contrast is usually appropriate as the next imaging study in the evaluation of chronic shoulder pain when radiographs are normal or inconclusive and labral pathology or shoulder instability is suspected.These procedures are equivalent alternatives (ie, only one procedure will be ordered to provide the clinical information to effectively manage the patient's care).
• Variant 5: Image-guided anesthetic +/-corticosteroid injection or MRI without IV contrast is usually appropriate as the next imaging study in the evaluation of chronic shoulder pain when radiographs are normal or inconclusive and adhesive capsulitis is suspected.These procedures are equivalent alternatives (ie, only one procedure will be ordered to provide the clinical information to effectively manage the patient's care).Although the panel did not agree on recommending US because there is insufficient medical literature to conclude whether these patients would benefit from the procedure, its use may be appropriate.
• Variant 6: US or Image-guided anesthetic +/-corticosteroid injection or MR arthrography or MRI without IV contrast is usually appropriate as the next imaging study in the evaluation of chronic shoulder pain when radiographs are normal or inconclusive and biceps tendon abnormality is suspected.These procedures are equivalent alternatives (ie, only one procedure will be ordered to provide the clinical information to effectively manage the patient's care).
• Variant 7: MRI without IV contrast is usually appropriate as the next imaging study in the evaluation of chronic shoulder pain when radiographs demonstrate osteoarthritis.Chronic Shoulder Pain • Variant 8: US or MR arthrography or MRI without IV contrast or CT arthrography is usually appropriate as the next imaging study in the evaluation of chronic shoulder pain with a history of prior rotator cuff repair when radiographs are normal or inconclusive and rotator cuff disorders or subacromial subdeltoid bursitis is suspected.These procedures are equivalent alternatives (ie, only one procedure will be ordered to provide the clinical information to effectively manage the patient's care).

Supporting Documents
The evidence table, literature search, and appendix for this topic are available at https://acsearch.acr.org/list.The appendix includes the strength of evidence assessment and the final rating round tabulations for each recommendation.
For additional information on the Appropriateness Criteria methodology and other supporting documents go to www.acr.org/ac.

Appropriateness Category Names and Definitions
Appropriateness Category Name Appropriateness Rating Appropriateness Category Definition Usually Appropriate 7, 8, or 9 The imaging procedure or treatment is indicated in the specified clinical scenarios at a favorable risk-benefit ratio for patients.
May Be Appropriate 4, 5, or 6 The imaging procedure or treatment may be indicated in the specified clinical scenarios as an alternative to imaging procedures or treatments with a more favorable risk-benefit ratio, or the risk-benefit ratio for patients is equivocal.
May Be Appropriate (Disagreement) 5 The individual ratings are too dispersed from the panel median.The different label provides transparency regarding the panel's recommendation."May be appropriate" is the rating category and a rating of 5 is assigned.
Usually Not Appropriate 1, 2, or 3 The imaging procedure or treatment is unlikely to be indicated in the specified clinical scenarios, or the risk-benefit ratio for patients is likely to be unfavorable.

Relative Radiation Level Information
Potential adverse health effects associated with radiation exposure are an important factor to consider when selecting the appropriate imaging procedure.Because there is a wide range of radiation exposures associated with different diagnostic procedures, a relative radiation level (RRL) indication has been included for each imaging examination.The RRLs are based on effective dose, which is a radiation dose quantity that is used to estimate population total radiation risk associated with an imaging procedure.Patients in the pediatric age group are at inherently higher risk from exposure, because of both organ sensitivity and longer life expectancy (relevant to the long latency that appears to accompany radiation exposure).For these reasons, the RRL dose estimate ranges for pediatric examinations are lower as compared with those specified for adults (see *RRL assignments for some of the examinations cannot be made, because the actual patient doses in these procedures vary as a function of a number of factors (eg, region of the body exposed to ionizing radiation, the imaging guidance that is used).The RRLs for these examinations are designated as "Varies."

Variant 7 :
Chronic shoulder pain.Initial radiographs demonstrate osteoarthritis.Next imaging study.

Variant 8 :
Chronic shoulder pain.History of prior rotator cuff repair.Suspect rotator cuff disorders or subacromial subdeltoid bursitis.Initial radiographs normal or inconclusive.Next imaging study.
Panel Vice-chair, University of Virginia Health System, Charlottesville, Virginia.b Panel Chair, Mayo Clinic Arizona, Phoenix, Arizona.c University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin.d Stanford University, Stanford, California, Primary care physician.e Mayo Clinic, Rochester, Minnesota.f Hospital for Special Surgery, New York, New York.g University of Nebraska Medical Center, Omaha, Nebraska; American College of Emergency Physicians.h The Centers for Advanced Orthopaedics, George Washington University, Washington, District of Columbia; American Academy of Orthopaedic Surgeons.i University of Michigan Health System, Ann Arbor, Michigan; Committee on Emergency Radiology-GSER.
a j Duke University Medical Center, Durham, North Carolina.k University of Missouri Health Care, Columbia, Missouri.l Cleveland Clinic, Cleveland, Ohio.m The University of Texas MD Anderson Cancer Center, Houston, Texas; Commission on Nuclear Medicine and Molecular Imaging.n Penn State Milton S. Hershey Medical Center, Hershey, Pennsylvania and Uniformed Services University of the Health Sciences, Bethesda, Maryland.o Specialty Chair, VA San Diego Healthcare System, San Diego, California.
[159]below).Additional information regarding radiation dose assessment for imaging examinations can be found in the ACR Appropriateness Criteria ® Radiation Dose Assessment Introduction document[159].Chronic Shoulder Pain