Baseline predictors of visual response and treatment burden after ranibizumab therapy in macular edema secondary to retinal vein occlusion.

BACKGROUND

Visual response to anti-vascular endothelial growth factor (anti-VEGF) therapy in macular edema secondary to retinal vein occlusion (RVO-ME) varies among patients, and baseline predictors remain incompletely defined. This study investigated baseline predictors of visual response and treatment burden after anti-VEGF therapy in eyes with RVO-ME.

METHODS

This retrospective observational study included 80 eyes from 80 patients with RVO-ME treated at a single center. Baseline clinical characteristics, optical coherence tomography (OCT) biomarkers, and hematologic inflammatory indices were collected. Decimal visual acuity was converted to approximate Early Treatment Diabetic Retinopathy Study (ETDRS) letter scores. Poor functional response was defined as a gain of ≤10 ETDRS letters at 1 month after the third anti-VEGF injection. A composite endpoint was used for sensitivity analysis. Treatment burden was assessed by the number of injections within 6 months.

RESULTS

Mean baseline and final ETDRS letter scores were 37.46 and 58.93, respectively, with a mean gain of 21.47 letters. Poor functional response occurred in 21 eyes (26.3%). After excluding sparse HRVO cases from regression modeling, CRVO was associated with poor functional response compared with BRVO (OR 6.81, 95% CI 1.17-39.74, p = 0.033), as were higher baseline ETDRS letter score per 10-letter increase (OR 1.85, 95% CI 1.20-2.84, p = 0.005) and male sex (OR 5.46, 95% CI 1.34-22.35, p = 0.018). Poor response was less frequent among patients with hypertension (OR 0.13, 95% CI 0.03-0.59, p = 0.008). Firth penalized logistic regression yielded similar results. In the continuous outcome model, CRVO was associated with lower final ETDRS letter score (β = -13.78 letters, 95% CI -19.62 to -7.93, p < 0.001). Conventional qualitative OCT biomarkers were not significantly associated with the primary gain-based endpoint, although CST showed stronger relevance in the composite endpoint analysis. In adjusted Poisson regression, CRVO showed a non-significant trend toward higher injection count.

CONCLUSION

In this small retrospective RVO-ME cohort treated with ranibizumab, CRVO was consistently associated with poorer short-term visual outcome. The association between baseline visual acuity and gain-based response appeared to be influenced by a ceiling effect. Conventional qualitative OCT biomarkers and hematologic inflammatory indices showed limited but potentially complementary value, and treatment-burden findings should be interpreted as exploratory.