Methods: This retrospective study included 89 treatment-naïve patients with branch RVO (BRVO) or central RVO (CRVO) followed for 12 months at a tertiary center. Patients received intravitreal bevacizumab, aflibercept, dexamethasone implant, or combination therapies. Baseline and follow-up best-corrected visual acuity (BCVA, logMAR) and central macular thickness (CMT) were recorded. Optical coherence tomography (OCT) parameters including serous macular detachment (SMD), hyperreflective foci (HRF), disorganization of the retinal inner layers (DRIL), and external limiting membrane (ELM) integrity were also assessed. Predictors of 12-month logMAR were analyzed using univariate and multivariate linear regression.

Results: The cohort included 65 BRVO (73.0%) and 24 CRVO (27.0%) patients (mean age 63.7 ± 12.1 years; 51.7% male). In univariate analysis, older age (p=0.002), CRVO type (p=0.004), baseline logMAR (p<0.001), baseline CMT (p=0.001), number of injections (p=0.01), HRF (p=0.023), DRIL (p=0.003), and ELM disruption (p=0.01) were significantly associated with poorer 12-month visual outcomes. Combination therapy with bevacizumab+dexamethasone was also linked to worse prognosis (p=0.009). In multivariate analysis (R²=0.467), only older age (p=0.003) and worse baseline BCVA (p=0.006) remained independent predictors, while baseline CMT, OCT biomarkers, treatment regimen, and occlusion type lost significance.

Conclusions: Age and baseline BCVA are the strongest independent predictors of long-term visual outcomes in RVO. Structural OCT features and treatment types show associations in univariate analyses but are not independent determinants after adjustment, underscoring the predominance of functional baseline status for prognosis. Keywords : Retinal vein occlusion, anti-VEGF therapy, macular edema, visual acuity