This study evaluates the diagnostic accuracy of anterior cruciate ligament (ACL) rupture using magnetic resonance imaging in the management of patients with symptomatic tibiofemoral instability. Published and unpublished databases including the Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE, AMED and CINAHL, as well as unpublished studies registers were searched for studies that compared the diagnostic findings of MRI to arthroscopic surgery for patients with a suspected ACL rupture in September 2010. For each study, sensitivity and specificity of complete ACL rupture were analysed as bivariate data to account for a possible negative correlation within studies, and pooled meta-analysis was performed. Fifty-three studies compared the diagnostic accuracy of MRI to arthroscopic findings for a total of 4,673 patients, 4,683 knees. Overall, the pooled sensitivity of MRI for detecting complete ACL rupture was 94.5% (95% confidence interval 0.92-0.96) and specificity of 95.3% (0.93-0.97). MRI has high accuracy for patients with complete ACL rupture. Due to the higher economic costs and increased complication risks of diagnostic arthroscopy, MRI is therefore recommended as the principal investigation following clinical examination, to diagnose ACL rupture.