Predictors of prognosis and treatment outcome in central retinal artery occlusion: local intra-arterial fibrinolysis vs. conservative treatment

Abstract

The study analyses patients' risk factors to determine prognostic and predictive factors in patients with acute central retinal artery occlusion (CRAO) treated in the randomized European Assessment Group for Lysis in the Eye (EAGLE) Study with local intra-arterial fibrinolysis (LIF) or conservative standard treatment (CST). These data could improve patient selection for either method. Post hoc statistical analysis of effects of risk factors on overall best corrected visual acuity (BCVA [logarithm of the minimum angle of resolution (logMAR)]) at baseline and month 1 (prognostic effect) and on the difference between outcome of CST and LIF (predictive effect) was conducted. Seventy two of 84 EAGLE datasets were included. Prognostic effect: Patients with coronary heart disease (CHD) presented worse BCVA at baseline (0.39 logMAR, p = 0.0097). Patients with time from occlusion to treatment < 12 h showed a trend to better vision gain at month 1 (-0.23 logMAR, p = 0.086), similarly smoking (-0.24 logMAR, p = 0.077). Predictive effect: Age (< 60 years favours LIF -0.54 logMAR; > 70 years favours CST 0.28 logMAR; interaction p = 0.070) and CHD (favours CST 0.44 logMAR; interaction p = 0.073) might be predictors of therapeutic outcome. There were no strong effects in multivariate analysis. CHD, time from occlusion to treatment and smoking influence BCVA at baseline and at month 1 (prognostic effect). Patients treated within 12 h are more likely to profit from treatment. In multivariate analysis, there is no clear trend to benefit from LIF even in patients with young age, no CHD and early treatment. Based on this preliminary report on a rather small sample size, we do not recommend LIF in CRAO patients.