Objectives To compare the clinical therapeutic effect of arthroscope and mini-open in treating rotator cuff impairment with Meta-analysi. Methods We searched the Cochrane Bone, Joint and Muscle Trauma Group Specialised Register (Jun 2007), the Cochrane Central Register of Controlled Trials (The Cochrane Library Issue 2, 2007), MEDLINE, EMBASE, and CBM, conference proceedings and reference lists of articles. Selection criteria: Randomized or comparative studies on all arthroscopic rotator cuff repairs and mini-open repairs. Results There were no randomized controlled trials (Level I) was found. Pooled results from all 12 trials showed that postoperative shoulder pain in all arthroscope group was statistically less than in the mini-open group (RR=0.94, 95%CI 0.28 to 1.60). Meanwhile, another evaluates outcomes such as ROM-Forward flexion (RR=0.17, 95%CI –0.10 to 0.45), patient’s satisfaction (RR=1.03, 95%CI 0.98 to 1.08), complication (RR=1.11, 95%CI 0.54 to 2.27), and shoulder functional score (RR=0.04, 95%CI –0.10 to 0.19) indicated no statistical difference in two groups. Conclusions According to Limited evidence, there are some findings as follows: comparing with mini-open treatment of rotator cuff impairment, all arthroscopic surgery can reduce the shoulder pain. Moreover, we found no statistical difference in shoulder functional score, ROM-Forward flexion, patient’s satisfaction and complication. Attention should be paid to outcome assessment in future trials.