Abstract Purpose Meniscus repair has gained increasing interest over the last two decades as loss of meniscus tissue predisposes to early onset knee arthritis. Although there are many reports of meniscus repair outcome in short-term studies, data on the long-term outcome of meniscus repair are still scarce. The purpose of this meta-analysis was to evaluate the overall failure rate of meniscus repair with a minimum follow-up of 5 years. Additionally, possible factors influencing meniscus repair outcome were assessed. Methods PubMed and Scopus were searched for studies of the last 20 years reporting on meniscus repair outcome with a minimum follow-up of 5 years. The study was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The search terms used for this study were ([meniscus OR meniscal] AND repair). Titles and abstracts were evaluated by two authors independently. Using meta package of R (version 3.6.2), random-effect models were performed to pool failure rates. Subgroup analyses were performed and effect estimates in form of an odds ratio with 95% CIs were established. Results In total, 12 studies with 864 patients were included. Degenerative tears were excluded in two studies and one study only included traumatic meniscus tears. Other studies did not state whether the cause of meniscus tear was degenerative or traumatic. Studies reporting meniscus repair outcome on root repairs, revision anterior cruciate ligament reconstruction, discoid menisci or ramp lesions were excluded. Revision surgery was used as failure definition in all included studies. The overall failure rate of meniscal repair at a mean follow-up of 86 months was 19.1%. There was no significant difference in meniscus repair outcome when performed in combination with anterior cruciate ligament reconstruction compared to isolated meniscus repair (18.7% vs. 28%; n.s.) or when performed on the lateral meniscus compared to the medial meniscus (19.5% vs. 24.4%; n.s.). There was no significant difference of meniscus repair outcome between vertical/longitudinal tears and bucket-handle tears (n.s.). Thirty-six percent of meniscus repair failures occur after the second postoperative year. The only significant finding was that inside-out repair results in a lower failure rate compared to all-inside repair (5.6% vs. 22.3%; p = 0.009) at 5 years. Conclusion The overall meniscus repair failure rate remains nineteen percent in long-term studies. The cause of failure is poorly documented, and it remains unclear whether failure of the meniscus repair itself or additional adjacent tears lead to revision surgery. Despite the given technical advantages of all-inside repair devices, this meta-analysis cannot demonstrate superior outcomes compared to inside-out or outside-in repair at 5 years. Level of evidence IV.