CT of the Shoulder for Tendon Tears
No, CT of the shoulder has virtually no usefulness in diagnosing tendon tears, including rotator cuff tears, and should not be used for this purpose. 1
Why CT Cannot Detect Tendon Tears
The American College of Radiology explicitly states that CT has virtually no usefulness in diagnosing common traumatic soft-tissue injuries such as rotator cuff tears, labroligamentous injuries, and muscle tears. 1 This is because:
- CT lacks adequate soft tissue contrast resolution to visualize tendon pathology, making it fundamentally unsuitable for evaluating rotator cuff or other tendon injuries 2
- CT is inferior to both MRI and ultrasound for diagnosing essentially all extra-articular traumatic soft-tissue injuries 1
- Even CT arthrography (with intra-articular contrast injection) is generally considered inferior to MRI in diagnosing rotator cuff tears and soft-tissue pathology 1
What CT Is Actually Good For
CT excels at evaluating bony pathology only:
- Gold standard for diagnosing and characterizing fractures, including complex fracture patterns 1
- Excellent for assessing bone loss, scapular fractures, and displaced fractures 1
- Useful for surgical planning when fractures are present 1
However, even for fractures, MRI has been shown to be equivalent to CT in diagnosing nondisplaced fractures that are typically missed on radiography 1
The Correct Imaging Approach for Suspected Tendon Tears
First-Line Imaging
- Plain radiographs (AP views in internal and external rotation plus axillary or scapula-Y view) should always be obtained first to exclude fractures and dislocations 3, 4
Definitive Soft Tissue Imaging
If radiographs are normal or nonspecific and tendon pathology is suspected:
- MRI without IV contrast is the optimal imaging modality for evaluating rotator cuff tears and other tendon pathology, with 94% sensitivity and 93% specificity for full-thickness tears 3, 5
- Ultrasound is comparable to MRI for detecting full-thickness rotator cuff tears (92% sensitivity, 93% specificity), though it has variable accuracy for partial-thickness tears and is operator-dependent 4, 5, 6
- MR arthrography should be reserved for cases where non-contrast MRI is equivocal for partial-thickness tears 3
Critical Pitfall to Avoid
Do not order CT when tendon pathology is the clinical concern. This wastes time, exposes patients to unnecessary radiation, and will not provide the diagnostic information needed. 1, 3 The rare exception is CT bursography for isolated bursal-side infraspinatus tears when MRI is normal but clinical suspicion remains high—however, this is an extremely specialized indication 7