The surgical treatment of adolescent idiopathic scoliosis has undergone significant evolution over the past 10 years. This has resulted from a better understanding of the natural history of the condition and the problems associated with the deformity, understanding the three-dimensional aspects of the deformity, and the development of new instrumentation used in treatment. It is quite likely that our knowledge in these areas will continue to advance in the years to come. The indications for surgical correction of spinal deformities is based on the natural history of the deformity itself and the potential consequences of the deformity for the patient in adult life. Current information regarding the natural history points to the potential consequences of living life with a significant visible deformity and pain. Pulmonary complications seem to result mainly from unusual deformities and early onset disease. In children and adolescents, surgery will often be done for curves when it is anticipated they will reach a magnitude expected to be troublesome to an adult. The primary goal for surgery remains the attainment of a solid arthrodesis and balance of the spine in both the sagittal and coronal planes with preservation of the maximum number of motion segments. Several additional factors, e.g., safety, cost, morbidity, are important in the surgical decision-making. Complications of spinal surgery remain a primary concern of patients and surgeons. Decreasing complications, e.g., neurologic deficit, infection, and instrument failure, is important. Also, important is the question of long-term changes related to the aging of the unfused spine below the fusion. This latter question and others that have been posed suggest the need for additional information regarding the long- term results of surgical intervention for various deformities. There is obviously a need for comparative analysis of long-term results of different interventions, but this will not be easy to achieve due to cost and funding constraints, difficulty in following patients over decades, the multiplicity of interventions used, and our evolving and changing treatments. We can expect to learn a great deal from improved evaluation of patients by prospective use of validated outcome instruments and performing focused assessment for patients treated in the past.