Perioperative risk of bleeding with antithrombotic agents in macular surgery: a national, prospective, multicentre study

{Purpose To compare the risk of haemorrhagic complications in elective macular surgery between patients with no antithrombotic (AT) treatment (defined as patients with no history of AT therapy or who discontinued AT therapy) and patients who continued AT treatment during the surgery. Methods E-case report forms were prospectively recorded in a database before vitreoretinal surgery and 1 month after. Data on patient characteristics, surgical techniques, haemorrhagic complications and antithrombotic status were collected. Patients with retinal detachment, proliferative diabetic retinopathy and previous retinal haemorrhage were excluded. Results A total of 748 procedures (single procedure in one eye per patient) were performed between January and May 2019. Among them, 202 patients (27.0%) were treated with antithrombotic therapy at the time of surgery: 19.5% with antiplatelet agents (n = 146), 6.3% with anticoagulants (n = 47) including 3.2% (n = 24) patients treated with novel oral anticoagulants, 0.8% (n = 6) with anticoagulants and antiplatelet agents, and 0.4% (n = 3) with heparin. Overall, 92 patients (12.3%) developed one or more haemorrhagic complications, of which 63 (11.5%) and 29 (14.4%) were in the non-AT and AT group, respectively. The multivariate logistic regression model showed no difference between AT treatment groups regarding ocular bleeding complications (odds ratio [OR] 1.2, 95% confidence interval (CI) [0.7-2.2]