Background: There is controversy over whether use of new oral anticoagulants (NOACs) associates with increased hemorrhage risk compared with non-NOAC. Meanwhile, determining which NOAC to use remains unclear. We aimed to summarize the evidence about NOACs in venous thromboembolism (VTE) prevention for patients with total hip and knee arthroplasty (THA and TKA).Methods: We searched RCTs assessing NOACs for VTE prophylaxis in adults undergoing THA and TKA in Medline, Embase, and Cochrane up to May 2021. Primary outcomes were VTE [included deep vein thrombosis (DVT) and pulmonary embolism (PE)], major VTE, and major bleeding. The rank probabilities of each treatment were summarized by the surface under the cumulative ranking curve area (SUCRA).Results: 25 RCTs with 42,994 patients were included. Compared with non-NOAC, NOACs were associated with a decreased risk of VTE (RR 0.68; 95% CI 0.55–0.84) and major VTE (RR = 0.52; 95% CI 0.35–0.76). Additionally, rivaroxaban, apixaban, and edoxaban but not dabigatran and betrixaban, did confer a higher efficacy compared with non-NOAC. None of the individual NOACs increased the risk of bleeding, while apixaban and betrixaban were even associated with a decreased risk of bleeding. In the comparison of different NOACs, rivaroxaban was associated with the greatest benefits in VTE (SUCRA = 79.6), DVT (SUCRA = 88.8), and major VTE (SUCRA = 89.9) prevention. Furthermore, subgroup analysis confirmed that NOACs associated with a higher efficacy tendency in patients with follow-up duration <60 days than follow-up duration ≥60 days.Conclusion: Evidence suggests that NOACs exert more benefits on VTE prophylaxis, and none of the individual NOACs increased hemorrhage compared with non-NOAC. Among various NOACs, rivaroxaban is recommended in patients with lower bleeding risk, and apixaban is recommended in patients with higher bleeding risk.Systematic Review Registration: [https://www.crd.york.ac.uk/prospero/], identifier [CRD42021266890].