Bleeding risks with novel oral anticoagulants especially rivaroxaban versus aspirin: a meta-analysis

BACKGROUND: This pairwise meta-analysis determines the difference in bleeding risks associated with the use of novel oral anticoagulants (NOACs) and aspirin. METHODS: PubMed, the Cochrane Library database, clinicaltrial.gov, and related studies were searched for randomized control trials (RCTs) comparing NOAC and aspirin published between January 1, 2000 and May 10, 2021. The primary endpoint was intracranial hemorrhage (ICH). RESULTS: Eleven studies involving 57,645 patients were included. Compared to aspirin, rivaroxaban (5 mg/day) had a similar risk of ICH, major bleeding, and fatal bleeding; rivaroxaban (10 mg/day) had higher risks of gastrointestinal hemorrhage (OR: 1.41; 95% CI: 1.03–1.94; P = 0.032; I(2) = 0%) and a similar risk of ICH, major bleeding, and fatal bleeding; and rivaroxaban (15–20 mg/day) had higher risks of ICH (OR: 3.21; 95% CI: 1.36–7.60; P = 0.008; I(2) = 0%), major bleeding (OR: 2.64; 95% CI: 1.68–4.16; P < 0.001; I(2) = 0%), and fatal bleeding (OR: 2.26; 95% CI: 1.25–4.08; P = 0.007; I(2) = 0%) and a similar risk of gastrointestinal hemorrhage. Bleeding outcomes between other NOACs (apixaban and dabigatran etexilate) and aspirin were not different. CONCLUSIONS: The bleeding risks associated with NOACs depend on drug type and dosage. For ≥15 mg/day of rivaroxaban, the risk of ICH was significantly higher than that with aspirin. However, further studies comparing dabigatran etexilate and apixaban versus aspirin are warranted to draw a definite conclusion. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12959-021-00322-6.