A modest increase in major bleeds has been found for rivaroxaban.
The anticoagulant rivaroxaban has a slightly increased risk of major bleeds compared with aspirin and with other NOACs, a systematic review has found with moderate to low certainty.
The study set out to confirm previous observational research and indirect evidence that rivaroxaban had the worst bleed risk out of the non-vitamin K oral anticoagulants.
Published in the Annals of Internal Medicine, the review and meta-analysis included nine studies comparing therapeutic-dose NOACs against single antiplatelet therapy for at least three months and reported bleeding events, with major bleeds and intracranial haemorrhage as the primary outcomes.
Among the combined 26,000 patients – who all had history of stroke, atrial fibrillation or venous thromboembolism – there were 566 cases of major bleeding (2.16%).
Apixaban and dabigatran had no clinically or statistically significant altered risk of major bleed compared with aspirin. Rivaroxaban, in two trials, had a 0.9 percentage point increase in risk, with a confidence interval of -0.1 to 3.7 percentage points.
There were 172 intracranial haemorrhages (0.66%), a low event rate that the authors said reduced the precision of their findings.
Again, apixaban and dabigatran were not significantly different from aspirin. Rivaroxaban, in two trials, showed a 0.3 percentage point increase in risk, with a very wide confidence interval of -0.1 to 79.7 percentage points.
The authors note that the small numbers and low precision may be hiding clinically more relevant differences, especially in patients with high bleeding risk, and that confounding was possible.
They said NOACs would be preferable to aspirin for most patients with “traditional” atrial fibrillation despite a small increase in bleeding risk, though the picture was less clear for transient postoperative AF and subclinical AF patients.
Though the anti-clotting benefits generally far outweigh the risks, “patients may consider the signal toward excess bleeding risk with rivaroxaban demonstrated in our study”.
