Risk and Predictors of Fellow-Eye Involvement after Unilateral Retinal Artery Occlusion.

PURPOSE: To determine the 5-year Kaplan-Meier (KM) estimated event probability, temporal pattern, and baseline predictors of fellow-eye retinal artery occlusion (RAO) after unilateral central RAO (CRAO) or branch RAO (BRAO).

DESIGN

Retrospective cohort study.

PARTICIPANTS

We identified 10 085 patients with a first-ever unilateral CRAO and 9574 patients with a first-ever unilateral BRAO within the TriNetX US Collaborative Network between January 1, 2015, and January 1, 2025. Patients with prior contralateral or bilateral RAO or systemic vasculitis were excluded.

METHODS

Fellow-eye RAO was defined as a new contralateral or bilateral RAO diagnosis 1 day to 5 years after the index event. Kaplan-Meier analysis estimated event probability. Baseline characteristics were compared between patients who did and did not develop fellow-eye RAO using χ2 and t tests with Bonferroni correction.

MAIN OUTCOME MEASURES: Five-year KM estimated event probability of fellow-eye RAO, time-to-event probability thresholds, and baseline characteristics associated with fellow-eye involvement.

RESULTS

Fellow-eye RAO developed in 409 of 10 085 patients with CRAO (4.06%) and 308 of 9574 patients with BRAO (3.22%), with CRAO carrying significantly higher risk (P = 0.002). Index RAO type predicted fellow-eye event type: 85.3% of patients with CRAO developed fellow-eye CRAO, and 82.1% of patients with BRAO developed fellow-eye BRAO (P < 0.0001). Among patients with CRAO, fellow-eye risk was higher for left-eye than for right-eye index events (4.56% vs. 3.60%, P = 0.018); no laterality difference was observed in BRAO. Fellow-eye events were front-loaded, with 2% event probability reached by days 31 to 83 in CRAO and days 124 to 168 in BRAO. After Bonferroni correction, in CRAO, significant predictors included cerebral infarction, carotid stenosis, hypertension, diabetes, diabetic retinopathy, and sleep apnea; in BRAO, significant predictors included retinal vasculitis, ocular hypertension, diabetic retinopathy, diabetes, and elevated hemoglobin A1c level.

CONCLUSIONS

Fellow-eye RAO occurs in ∼1 in 25 patients with CRAO and 1 in 31 patients with BRAO within 5 years. The concordance between index and fellow-eye RAO subtype and divergent systemic versus ocular risk profiles supports distinct pathophysiologic pathways. These findings provide counseling data and support close follow-up during the first several months after unilateral RAO.

FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.