This study retrospectively analyzed 1007 diagnostic shoulder arthroscopies. The study included 72 patients with an arthroscopically verified pulley lesion as the main pathologic finding. Epidemiologic data and arthroscopic findings were evaluated in all patients, and 53 were clinically examined with the Constant score after a minimum follow-up of 2 years. We observed an incidence of 7.1% for pulley lesions. An isolated rupture of the superior glenohumeral ligament (SGHL) was seen in 53 patients (73.6%) and a combined partial articular-side tear of the rotator cuff adjacent to rotator interval in 19 (26.4%). Thirty-one patients (43%) had a history of trauma, whereas 41 (57%) had none. Overall, the mean postoperative Constant score adjusted for age and gender was 80.1% (range, 47%-135%). Patients with a SGHL lesion only (85.7%) exhibited a significant (P = .047) higher age- and gender-adjusted Constant score compared with patients with a combined partial surface tendon tear (73.1%). Our epidemiologic data accentuate the need for careful evaluation of the superolateral aspect of rotator interval to avoid underdiagnosis of pulley lesions at shoulder arthroscopy. Our findings provide evidence that the clinical outcome of isolated SGHL lesions is better compared with combined partial articular-side rotator cuff tear. With respect to the progressive pathologic process of pulley lesions, we recommend an early surgical treatment.