Diagnosis and Management of Partial Thickness Rotator Cuff Tears

A Comprehensive Review

Kevin D. Plancher, MD, MPH; Jaya Shanmugam, MD; Karen Briggs, MPH; Stephanie C. Petterson, MPT, PhD


J Am Acad Orthop Surg. 2021;29(24):1031-1043. 

In This Article


Initial evaluation should include plain radiographs, including true AP, axillary, and scapular-Y views. Although plain imaging is often nonspecific, it can assess for fracture, glenohumeral osteoarthritis, acromial spurs, or other sources of pain. A reduction in the acromiohumeral distance is usually indicative of a FTT (normal 9 to 14 mm).

Ultrasonography (US) has been increasingly used as an imaging modality due to its availability, portability, cost-effectiveness, and dynamic visualization. Confirmatory ultrasonography findings of PRCTs include a focal, contour tendon defect, a mixed hypohyperechoic linear band, or a linear band of anechoic appearance (Figure 1). Although ultrasonography is reliable in diagnosing FTTs, distinguishing PRCTs from tendon scarring and small FTTs can be difficult. The reported accuracy, sensitivity, and specificity of US in detecting PRCTs are 87%, 66.7%, and 93.5%, respectively.[18]

Figure 1.

Ultrasound image showing a small articular-sided partial thickness cuff tear.

MRI provides valuable information on the anatomy and structural integrity of the RC, including tear location, size, and muscle atrophy, as well as other soft-tissue injuries (eg, labral tears and biceps tendon pathology). MRI findings indicative of a PRCT include increased signal intensity at the RC insertion onto the greater tuberosity on either the bursal or articular surfaces or within the tendon substance on T2-weighted, fat-suppressed images matching fluid signal (Figures 2–4). MRI is invaluable in diagnosing intratendinous tears that are not evident on arthroscopic evaluation because only the outside surface of the tendon can be visualized; however, they are often missed because of patient positioning. When placed in the MRI scanner with the arm positioned at the side, the layers of the RC are compressed and do not allow for visualization of interstitial tears.

Figure 2.

T2-weighted fat suppressed image showing a >50% thickness articular-sided partial thickness rotator cuff tear in a right shoulder of a 75-year-old woman. Copyright K. Plancher, MD, MPH

Figure 3.

T2-weighted fat-suppressed image showing a bursal-sided partial thickness rotator cuff tear involving >50% thickness in a right shoulder of a 56-year-old man. Copyright K. Plancher, MD, MPH

Figure 4.

T-2 weighted image showing an intrasubstance tear of the supraspinatus tendon in a right shoulder. Copyright K. Plancher, MD, MPH

A systematic review of 44 studies reported a sensitivity and specificity of MRI in detecting PRCTs of 80% and 95%, respectively.[19] However, Brockmeyer et al[20] did not demonstrate the same degree of accuracy with MRI (sensitivity 51.6, specificity 77.2, positive predictive value 41.3%, and negative predictive value 83.7%). Although some suggest similar sensitivity and specificity between MRI and ultrasonography, a meta-analysis of 144 studies showed MRI to have a higher sensitivity and superiority index compared with ultrasonography.[21]

Magnetic resonance arthrography uses the injection of gadolinium contrast into the joint for evaluation. Articular-sided tears are visualized as tendon discontinuity with contrast extruding from the articular side on T2-weighted images (Figure 5). Bursal tears are seen as focal hyperintense lesions on fat-saturated intermediate-weighted images, extending to the subacromial bursa.[22] Magnetic resonance arthrography has superior diagnostic value in terms of sensitivity and specificity for articular-sided PRCTs compared with ultrasonography and MRI, although similar diagnostic accuracy has not been found for bursal-sided tears.[23]

Figure 5.

MRA of the right shoulder showing an articular-sided partial thickness rotator cuff tear in a 47-year-old woman. Copyright K. Plancher, MD, MPH