The purpose of this study was to compare long-term results after surgical and conservative primary treatment of first-time traumatic anterior shoulder dislocation. Arthroscopic diagnosis after first-time traumatic anterior shoulder dislocation was performed, and in cases of a Baker type 1, 2, or 3 lesion, patients were randomized either to conservative treatment with a fixed sling for 1 week followed by a rehabilitation program or to open repair with a similar rehabilitation program. In this study 76 patients (14 female and 62 male patients), aged 15 to 39 years, were randomized to surgical repair (n = 37) or conservative treatment (n = 39). Of the patients, 6.6% had Baker type 1 lesions, 13.2% had type 2 lesions, and 80.3% had type 3 lesions. After a minimum of 2 years' follow-up, 56% had recurrence after conservative treatment and 3% after open repair (P < .005). Among nondislocators, 39% in the conservative group and 7% in the repair group had a positive apprehension test. When evaluated after 10 years by use of the Oxford self-assessment score, 72% of patients in the surgical group had good or excellent results. Of the conservatively treated patients, 75% had unsatisfactory results because of recurrence, instability, and pain or stiffness. Arthroscopic evaluation after first-time anterior shoulder dislocation revealed a Baker type 2 or 3 lesion in 93.5% of patients. At 2 years' follow-up, 21 (54%) of the conservatively treated patients had recurrence, as compared with 1 patient with recurrence (3%) after open surgical repair. After 8 years, a further 3 patients in the conservatively treated group had redislocations, 1 had subjective instability, and 4 had pain or stiffness, resulting in 74% having unsatisfactory results according to the Oxford score. Of the patients who had surgical repair, 72% had good or excellent results after 10 years. Because open repair produces superior results compared with conservative treatment, we recommend that the surgeon consider performing primary repair in active patients to reduce the risk of recurrence. Level I, high-quality prospective, randomized controlled trial.