The purpose of this study was to evaluate the effect of surgical reinsertion of the medial patellofemoral ligament (MPFL) to the adductor tubercle compared with conservative treatment in patients with primary dislocation of the patella. Eighty patients with primary patella dislocation were included in the study. Delayed arthroscopy (mean, 50 days after injury) was performed to assess cartilage injury and status. During arthroscopy, patients were randomized to surgical reinsertion of the MPFL or to conservative treatment. Conservative treatment was use of a brace with 0 degrees to 20 degrees motion for the first 2 weeks after dislocation. Patients were followed up for 2 years. The incidence of recurrent patella dislocation was recorded along with the clinical Kujala scores and Knee Injury and Osteoarthritis Outcome Scores. The redislocation rates were 17% and 20% in the operative and conservative treatment groups, respectively (P = not significant). Kujala scores were 85 and 78 in the operative and conservative treatment groups, respectively (P = .07). The patella stability subscore was significantly higher in the operative group. No difference in Knee Injury and Osteoarthritis Outcome Scores was found. Delayed primary repair of the MPFL by use of an anchor-based reattachment to the adductor tubercle without vastus medialis obliquus repair after primary patella dislocation does not reduce the risk of redislocation nor does it produce any significantly better subjective functional outcome based on the Kujala knee score. Only the specific subjective patella stability score was improved by MPFL repair compared with conservative treatment. Level I, therapeutic randomized controlled trial.