Effect of direct oral anticoagulants on bleeding during and after cataract surgery

**Background:** Direct oral anticoagulants (DOACs) are increasingly used for thromboprophylaxis in conditions like atrial fibrillation, but data on their safety during cataract surgery are limited. Phacoemulsification is considered a low-bleeding-risk procedure, and guidelines recommend continuing warfarin, yet evidence for DOACs is sparse. This study aimed to assess the risk of intraoperative and postoperative ocular bleeding in patients continuing DOACs during cataract surgery compared to those not on anticoagulation.

**Methods:** This retrospective review included consecutive patients aged >50 years undergoing phacoemulsification with IOL implantation at Meir Medical Center from January 2017 to July 2019. The study group comprised 40 patients (56 eyes) on uninterrupted DOACs (dabigatran, rivaroxaban, or apixaban). A control group of 120 age- and gender-matched patients (172 eyes) without anticoagulation was selected in a 3:1 ratio. Exclusion criteria included general or retrobulbar anesthesia, use of other anticoagulants/antiaggregants, traumatic cataract, posterior synechiae, rubeosis iridis, subluxated lens, or combined ocular surgery. Data on demographics, systemic/ocular comorbidities, CHA2DS2-VASc score, renal function, DOAC type, surgical risk factors, anesthesia type, visual acuity, and complications were collected. All surgeries used topical anesthesia with intracameral lidocaine, clear corneal incisions (2.2–2.4 mm), and phacoemulsification with the Infiniti Vision System. Postoperative treatment included topical steroids and antibiotics. Patients were examined at 1 day and 7–10 days postoperatively. Primary outcomes were intraoperative and postoperative hemorrhagic complications. Statistical analysis used t-tests, chi-square/Fisher's exact tests, and Mann-Whitney tests, with p≤0.05 considered significant.

**Key Results:** The study group had a mean age of 77.3 years (60% male), with 57.5% on apixaban, 27.5% on rivaroxaban, and 15% on dabigatran; 80% had atrial fibrillation. The control group had a mean age of 77.2 years (60% male). Systemic comorbidities (hypertension, ischemic heart disease, heart failure, cerebrovascular disease) and CHA2DS2-VASc scores were significantly higher in the study group (p<0.05). Median preoperative visual acuity was 6/15 in both groups. Ocular comorbidities were similar, but alpha-blocker treatment was more common in the study group (35.7% vs. 15.1%, p<0.01). Intraoperative hyphema occurred in 1 eye (1.8%) of the study group and 2 eyes (1.2%) of controls (p=0.57). Postoperative hyphema was noted in 1 eye (1.8%) vs. 3 eyes (1.7%) (p=0.72). Total hemorrhagic complications (including eyelid hematoma, subconjunctival hemorrhage, hyphema, vitreous hemorrhage, retinal hemorrhage) occurred in 2 eyes (3.6%) of the study group and 5 eyes (2.9%) of controls (p=0.57). No thromboembolic events occurred in either group. Intraoperative floppy iris syndrome (IFIS) showed a nonsignificant trend toward higher incidence in the study group (14.3% vs. 6.4%, p=0.06). Postoperative visual acuity was similar (median logMAR 0.30 vs. 0.22, p=0.83).

**Clinical Implications:** This study provides evidence that continuing DOACs during cataract surgery does not significantly increase hemorrhagic complications compared to no anticoagulation, with a low overall bleeding rate (3.6% vs. 2.9%). The findings align with guidelines recommending uninterrupted anticoagulation for low-bleeding-risk procedures, though ophthalmology-specific guidelines are lacking. The higher rate of IFIS in the DOAC group likely reflects greater alpha-blocker use rather than a direct drug effect. Limitations include the retrospective design, small sample size, and lack of comparison with DOAC withholding. Nonetheless, the results support the safety of uninterrupted DOAC therapy, balancing thrombotic risk against minimal bleeding risk. Larger prospective studies are needed to establish definitive guidelines.