The most commonly recommended treatment for multidirectional instability (MDI) is exercise-based management, followed by surgery if exercise management fails. We aimed to evaluate the effectiveness of exercise-based management compared with surgery in patients with MDI. Secondary aims were to identify effective protocols and any adverse effects associated with exercise or surgery.The Cochrane database of systematic reviews, MEDLINE, EMBASE, CINAHL, PEDro, Current Contents, AMED, AMI, Ausport and Clinical Trials Registers were searched for published and unpublished studies from inception date to August 2014, using the key words multidirectional instability, and glenohumeral and exercise and surgery. Selection criteria included all study designs (except case reports and case series) and participants with MDI where both exercise-based management and surgery were compared in the same study. Inclusion criteria were not limited by outcomes. Risk of bias was assessed using Cochrane criteria. The GRADE approach was used to synthesise the evidence.4 non-randomised studies met the inclusion criteria. Risk of bias was high in all 4 studies. GRADE assessment revealed very low-quality evidence that surgery was superior to exercise therapy for impairment outcome measures, and exercise was favoured over surgery for patient-reported outcome measures.The effect of exercise-based management compared with surgery for MDI is difficult to determine due to participant heterogeneity and a high level of bias across included studies.