BACKGROUND: Conventional reconstruction of the anterior cruciate ligament (ACL) is associated with a 15% failure rate. Computer-assisted navigation systems (CANS) have been developed to improve the accuracy of tunnel positioning. HYPOTHESIS: The use of a CANS for ACL reconstruction decreases the rate of failure, defined as IKDC grade C or D, compared to conventional ACL reconstruction. MATERIALS AND METHODS: This prospective multicentre observational non-randomised open study compared two groups of patients requiring arthroscopic ACL reconstruction: one group was managed with a CANS and the other (control group) without a CANS. The primary evaluation criterion was based on the subjective and objective IKDC scores. Inclusion criteria were age older than 18 years and first ACL reconstruction procedure using autologous semitendinosus and gracilis tendons or an autologous bone-patellar tendon-bone graft. Of the 272 included patients, 214 were analysed; 100 were in the control group and 114 in the CANS group. RESULTS: No significant between-group differences were found for the fraction of patients having an IKDC grade A or B (P=0.953), the subjective IKDC score (P=0.77), differential knee laxity at 150 N (1.38 ± 1.79 mm in the control group and 1.77 ± 2.06 mmin the CANS group, P=0.384), graft-type, or graft positioning. DISCUSSION: Our results establish the large-scale feasibility of computer-assisted navigation for ACL reconstruction. However, the main outcomes at 1 year showed no significant differences between patients managed with and without computer-assisted navigation.