STUDY DESIGN: A retrospective study. OBJECTIVES: To determine the indications of fusing the proximal thoracic curve when treating an idiopathic thoracic scoliosis with segmental instrumentation. SUMMARY OF BACKGROUND DATA: Failure to recognize a significant proximal thoracic curve often results in postoperative shoulder asymmetry due to relative overcorrection of the lower thoracic curve. With segmental instrumentation that enhances the correction of the instrumented curve, the double thoracic curve pattern that needs fusion of both the proximal and the distal thoracic curves needs to be redefined. MATERIALS AND METHODS: Forty thoracic AIS patients with a right lower thoracic curve > 40 and a left proximal thoracic curve > 25 treated by segmental pedicle screw instrumentation were analyzed after a minimum follow up of 2 years. RESULTS: Of the 40 patients, 18 were treated by fusion of both the proximal and the distal curves while 22 were treated by fusion of the distal curve only. The postoperative shoulder height difference(SHD, mm) was 0.9 x preoperative SHD(mm) + 5.3 for the both curve fusion and 0.6 x preoperative SHD(mm) + 12 for the distal curve fusion(linear regression), showing that proximal thoracic curve fusion improved the SHD when the left shoulder was level with or higher than the right. CONCLUSIONS: An idiopathic thoracic scoliosis with a proximal thoracic curve > 25 and level or elevated left shoulder should be considered a double thoracic curve pattern and treated by both the proximal and the distal curve fusion when using a segmental instrumentation.