Fifty-eight patients who sustained traumatic small rotator cuff tears, mainly of the supraspinatus tendon, were randomized to receive arthroscopic rotator cuff repair (n=32) plus physical therapy or physical therapy alone (n=26). The primary outcome of interest was pain and function evaluated using the Constant-Murley (CM) score (i.e. total pain, activities of daily living, internal rotation, external rotation, flexion, abduction, and strength) at 1-year post-treatment. Secondary outcomes of interest included the Western Ontario Rotator Cuff (WORC) index, pain evaluated using a Numeric Rating Scale (NRS), the Euro quality-of-life on visual analog scale (EQ-VAS). Furthermore, MRI characteristics regarding healing, tear progression, atrophy, and fatty infiltration were also evaluated at 12-months post-treatment. Results demonstrated that for the primary outcome of interest, none of the CM scores were statistically significantly different between the two treatment groups (p>0.05 for all). The median sagittal and coronal tear sizes were statistically significantly smaller in the surgical repair group compared to the physiotherapy group (p<0.001 for both). All other outcomes were similar between the two groups (p>0.05 for all).