Since February 2000, arthroscopic rotator cuff repairs have been performed. We evaluated 73 shoulders of 75patients followed-up for more than 6 months. The average age at the operation was 64 years old. There were 64full-thickness tears. Small and medium tears were observed in 47 shoulders. Large and massive tears were recognized in 17 shoulders. There were 11 partial thickness tears,6 bursal side and 5 joint side tears. We used anchors in 74 shoulders. Anchor placement was a single row in all shoulders. In 4 shoulders of large and massive tears, partial cuff repairs were performed. In 2 shoulders of massive tears, arthroscopic tensor fascia lata grafts were done. We evaluated the results using the Japanese Orthopaedic Association Shoulder score. JOA score improved significantly. Excellent and good results were obtained in 91 %. In full-thickness tears, small and massive tears had better results than large and massive tears significantly. In partial-thickness tears, bursal side tears had better results compared with joint side tears. About 3 months post-operative MRI T2findings were evaluated by Sugaya's classification. In type 1,2,3, cuff continuity was preserved. Type 4 was minor cuff discontinuity. Type 5 was major cuff discontinuity. Type 1,2,3 were recognized in 66% and type 4and 5 were observed in 34%. In post-operative JOA score, type 1,2,3 had better results than type 4 and 5significantly. Type 4 and 5 were observed in 28% of small and medium tears, in 65% of large and massive tears and 9% of partial thickness tears. There were many problems in large and massive tears. Tendon retraction, tendon substance loss, tendon quality, muscle contraction, muscle quality. We could not resolve several problems surgically. The best indication was for small to medium tears and partial-thickness tears. Good indication was for large to massive tears.
