This meta-analysis of data from phase 3 clinical trials examined the risk of intraocular bleeding with novel oral anticoagulants compared with warfarin.
Study design
Investigators included 12 phase 3 randomized clinical trials comprising 102,627 patients with atrial fibrillation or venous thromboembolism. Studies were eligible for review if patients were randomized to either warfarin or a novel anticoagulant (dabigatran, rivaroxaban, apixaban or edoxaban), and if intraocular bleeding was recorded as a predefined outcome, or an adverse event.
Outcomes
Patients who were randomized to novel oral anticoagulants were associated with a 22% relative reduction in intraocular bleeding compared with those on warfarin (RR 0.78). The reduced risk was similar regardless of the indication for anticoagulation (atrial fibrillation or venous thromboembolism) or the type of novel anticoagulant.
Limitations
Because the overall incidence of intraocular bleeding is low, it necessitates pooling of data. Clinical trials have a relatively short follow-up, and further research is needed to assess long-term risk or benefit. In addition, the authors did not have access to individual patient data, and were therefore unable to perform further subgroup analysis.
Clinical significance
These data are important for patients who are at higher risk of spontaneous intraocular bleeding, such as those with hypertension and exudative AMD. Though 1 study found that vitrectomy can be safely performed without discontinuation of novel anticoagulants, they still may need to be halted 48 hours prior to complex oculoplastics cases such as dacryocystorhinostomy, deep orbital and extensive eyelid surgeries.
Retina specialists won’t likely change how they manage patients as a result of these findings. But these data suggest that further communication with the patient's cardiologist or primary care physician on the associated risks of anticoagulants may be prudent.