Thrombolysis for central retinal artery occlusion: An individual participant-level meta-analysis

Background: Whether thrombolysis improves outcomes in non-arteritic central retinal artery occlusion (naCRAO) is uncertain. We aimed to evaluate the rate of visual recovery after intra-venous thrombolysis (IVT) or intra-arterial thrombolysis (IAT) administration of tissue plasminogen activator (tPA) or urokinase among patients with naCRAO and explore the parameters affecting the final visual acuity (VA). Aim: We systematically searched six databases. Logarithm of the minimum angle of resolution (logMAR) and VA of ⩾20/100 were used to quantify visual recovery. To explore the role of other factors on visual recovery, we defined two models for studies with aggregated data (designs 1 and 2) and 16 models for individual participant data (IPD, models 1–16). Summary of Review: We included data from 771 patients out of 72 publications in nine languages. Visual improvement for ⩾0.3 logMAR was reported in 74.3% of patients who received IVT-tPA within 4.5 h (CI: 60.9–86.0%; unadjusted rate: 73.2%) and 60.0% of those who received IAT-tPA within 24 h (CI: 49.1–70.5%; unadjusted rate: 59.6%). VA of ⩾20/100 was observed among 39.0% of patients after IVT-tPA within 4.5 h and 21.9% of those with IAT-tPA within 24 h. IPD models highlighted the association between improved visual outcomes and VA at presentation, at least 2 weeks follow-up before reporting the final VA, antiplatelet therapy, and shorter symptom onset to thrombolysis window. Conclusion: Early thrombolytic therapy with tPA is associated with enhanced visual recovery in naCRAO. Future studies should refine the optimum time window for thrombolysis in naCRAO.