Much debate exists over the value of exercise therapy for treating adolescent idiopathic scoliosis (AIS). This study aims to address the current evidence. An extensive search was carried out using the common medical databases, limiting results to clinical trials in English involving humans with defined outcome measures. 155 papers were identified and after applying strict inclusion criteria 12 papers remained for further analysis. These included 9 prospective cohort trials, 2 retrospective cohort trials and one case series. No randomised controlled trials were identified. Although all of the papers concluded an improvement in scoliotic curve after exercise therapy, not one of the papers had reliable methods or results to validate their conclusions. Identified shortcomings included, poor compliance with outpatient exercise regimes with no clear indication of who assessed for curve improvement, how it was assessed or what experience they had, nor was observer error for Cobb angle measurement taken into account. Additionally only a few of the studies had sound statistical analysis and no study could comment on whether the improvements seen were maintained after the exercise regime. Four previous systematic reviews have been performed finding favourable results for exercise therapy, but these were written by authors involved in the original research, adding significant reviewer bias. This systematic review has revealed only poor and low level evidence supporting the use of exercise therapy for treating AIS. Well designed controlled trials with randomisation are required to validate exercise therapy as an effective treatment option and as an appropriate use of NHS funds.