Bleeding Risks With Non-Vitamin K Oral Anticoagulants Versus Single Antiplatelet Therapy : A Systematic Review and Meta-analysis of Randomized Trials.

Wang, Michael Ke; Baskaran, Geethan; Razeghi, Ghazal; Ma, Richard; Park, Louis S; Tannu, Manasi; Devereaux, P J; McIntyre, William F et al. · Ann Intern Med · 2025

meta_analysis · Level I

Where this comes from

Abstract

In several settings, therapeutic-dose non-vitamin K oral anticoagulants (NOACs) are superior to aspirin for the prevention of arterial and venous thromboembolism. To estimate differences in bleeding risks between NOACs and single antiplatelet therapy. MEDLINE, Embase, Cochrane Central Register of Controlled Trials, and ClinicalTrials.gov from inception to June 2024 without language restrictions. Randomized controlled trials (RCTs) comparing therapeutic-dose NOACs versus single antiplatelet therapy, with a minimum treatment duration of 3 months. Data extraction was done independently and in duplicate. Nine RCTs with 26 224 participants were included. All studies used aspirin as antiplatelet therapy. Compared with aspirin, apixaban had similar rates of major bleeding (risk difference [RD], 0.0 percentage point [95% CI, -1.3 to 2.6 percentage points]; 5 trials) and intracranial hemorrhage (RD, -0.2 percentage point [CI, -0.6 to 1.4 percentage points]; 5 trials). Compared with aspirin, dabigatran had similar rates of major bleeding (RD, 0.5 percentage point [CI, -2.1 to 19.6 percentage points]; 2 trials) and intracranial hemorrhage (RD, 0.0 percentage point [CI, -1.1 to 24.5 percentage points]; 2 trials). Compared with aspirin, rivaroxaban had higher rates of major bleeding (RD, 0.9 percentage point [CI, -0.1 to 3.7 percentage points]; 2 trials) and intracranial hemorrhage (RD, 0.3 percentage point [CI, -0.1 to 79.7 percentage points]; 2 trials). The evidence certainty ranged from low to moderate. Confidence intervals were wide and included the possibility of a null effect. In this systematic review of RCTs, rates of major bleeding for therapeutic-dose apixaban and dabigatran were similar to those for low-dose aspirin, whereas rates were higher for rivaroxaban. None. (PROSPERO: CRD42024553683).

Medical subject headings