Objective
To determine the association between the type and severity of rotator cuff tear and ultrasonographic findings. Method: Total 91 cases with rotator cuff tear were classified into partial- or full-thickness tear according to arthrographic and ultrasonographic findings. Partial-thickness tear was divided into the tear on the bursal or articular side. Full-thickness tear was divided into small (2 cm＜) or large (2 cm≧) tear by the length of retraction. Combined findings which included biceps tenosynovitis, subacromial-subdeltoid (SASD) bursitis, acromio-clavicular (AC) joint osteoarthritis and the cortical irregularity of greater tubercle (GT) were evaluated by ultrasonography. Results: Sixty five cases (71.9%) were full-thickness tear and 26 cases (28.1%) were partial-thickness tear. Forty two cases (46.1%) were SASD bursitis, 33 cases (36.2%) were biceps tenosynovitis, 24 cases (26.3%) were cortical irregularity of GT, and 26 cases (28.5%) were AC joint osteoarthritis. The incidence of SASD bursitis was higher in full- thickness tear than partial-thickness tear. The SASD bursitis and cortical irregularity of GT were more frequently shown in large tear than small tear. Conclusion: SASD bursitis showed higher association with full-thickness tear than partial-thickness tear. The length of retraction was more severe when SASD bursitis or cortical irregularity of GT was associated with full-thickness tear. (J Korean Acad Rehab Med 2005; 29: 291-296)