PURPOSE: To evaluate the results from arthroscopic repair versus debridement for full-thickness tears of the upper subscapularis tendon. MATERIALS AND METHODS: Ninety-nine patient outcomes were evaluated and compared prospectively. Sixty-nine patients with full-thickness tears of the upper subscapularis tendon underwent arthroscopic repair (group I) and thirty patients underwent simple debridement (group II) between May 2003 and October 2007. In all patients, the tear was localized to the superior one third of the upper subscapularis tendon. The results of the treatment were assessed by evaluating the UCLA, ASES, and VAS for pain and internal rotation strength scores before and after the operation. RESULTS: In groups I and II, UCLA, ASES, VAS, and internal rotation muscle power (perfect score = 5) scores were improved after surgery. In comparing group I and group II, the UCLA and VAS scores were not significantly different (p>0.05), while the ASES and internal rotation strength scores were significantly different (p<0.05). CONCLUSION: The arthroscopic repair of full-thickness tears of the upper subscapularis tendon was a better surgical method than simple debridement.