No, CT is NOT sufficient to rule out rotator cuff pathology in this patient
CT shoulder without contrast receives an appropriateness rating of only 1 (usually not appropriate) for evaluating suspected rotator cuff tears, and you should proceed with MRI or ultrasound imaging. 1
Why CT Fails for Rotator Cuff Evaluation
The ACR Appropriateness Criteria are unequivocal on this point: CT without contrast is rated 1 out of 9 (usually not appropriate) for evaluating rotator cuff pathology across multiple clinical scenarios. 1 This rating means CT should essentially never be used for this purpose.
The evidence demonstrates CT's critical limitations:
CT significantly underestimates rotator cuff tears: In patients with glenohumeral osteoarthritis, CT showed only 20% sensitivity for detecting full-thickness rotator cuff tears compared to MRI's gold standard. 2
CT misses partial-thickness tears: Even CT arthrography (which is superior to plain CT) has significantly poorer performance for partial-thickness tears compared to MRI. 3, 4
CT underestimates muscle pathology: CT significantly underestimates fatty infiltration in the supraspinatus (p=.003), infraspinatus (p<.001), and subscapularis (p=.0182), as well as muscle atrophy in the supraspinatus (p=.0023). 2 These findings are critical for surgical planning and prognosis.
What You Should Order Next
Order MRI shoulder without contrast (rated 9/9 - usually appropriate) or ultrasound shoulder (rated 9/9 - usually appropriate). 1, 5
MRI is preferred because:
- High sensitivity and specificity for both full-thickness and partial-thickness rotator cuff tears 1
- Accurately identifies tendon retraction, muscle atrophy, and fatty infiltration—critical factors for determining conservative vs. operative management 1
- Can distinguish between tendinosis, partial-thickness tears, and full-thickness tears 1
- Provides comprehensive evaluation of other shoulder pathology (labral tears, biceps tendon, bursitis) 1
Ultrasound is an acceptable alternative if:
- Local expertise is available (operator-dependent) 1
- Can evaluate rotator cuff tendons, acromioclavicular joint, and biceps tendon 1
- Limited in evaluating deep shoulder structures and bone marrow 1
Clinical Context for This 85-Year-Old Patient
The minimal relief from steroid injection is actually diagnostically significant. Moderate evidence supports that corticosteroid injection provides short-term improvement in pain and function for rotator cuff pathology. 6 The lack of response suggests either:
- A structural tear requiring different management (not just inflammation/tendinopathy)
- Alternative pathology that CT cannot adequately evaluate
At 85 years old, this patient is at high risk for rotator cuff tears, and strong evidence shows that older age is associated with higher failure rates and poorer outcomes after rotator cuff repair. 6 This makes accurate diagnosis even more critical for appropriate treatment planning—whether conservative management, surgical repair, or alternative interventions.
Common Pitfall to Avoid
Do not assume CT "ruled out" a rotator cuff tear simply because one was performed. CT is useful for characterizing fractures and bony anatomy but provides inadequate soft tissue visualization. 1 The 2025 ACR guidelines explicitly state there is "no relevant literature to support the use of CT shoulder without IV contrast as the next study in assessment of suspected rotator cuff tear." 5
When CT Arthrography Might Be Considered
If MRI is absolutely contraindicated (pacemaker, severe claustrophobia unresponsive to sedation, metallic foreign body) and ultrasound expertise is unavailable, CT arthrography receives a rating of 5/9 (may be appropriate) as a last resort. 1 However, even CT arthrography has significantly lower sensitivity for partial-thickness tears compared to MRI. 3, 4