Time-, Location-, and Layer-Specific Chorioretinal Biomarkers in a Translational Large-Animal Porcine Model of Chorioretinal Ischemia

Purpose: Conditions associated with complete or partial ocular ischemia, such as central retinal artery occlusion (CRAO), are vision-threatening emergencies. Therapeutic interventions are highly time-dependent, yet the exact onset of ischemia is often uncertain. Optical coherence tomography (OCT) has emerged as a potential tool for objective staging; however, quantitative, layer-resolved OCT data within the earliest phase of retinal ischemia in clinically relevant large-animal models are lacking. Methods: Ten eyes from 10 pigs were included. Physiological retinal perfusion was confirmed using fluorescein-, indocyanine-green-, and OCT angiography (OCTA). Baseline chorioretinal morphology was assessed by OCT. Global retinal ischemia was induced by controlled cardiac arrest. Electrophysiological response of the retina after ischemia was measured using electroretinography (ERG) and serial OCT imaging was performed at intervals of 2.5 to 30 minutes for up to 270 minutes postmortem. Layer- and region-specific changes were quantified using dedicated image analysis software. Results: ERG response was abolished within 2 minutes after cardiac arrest and structural retinal changes were immediately detectable and followed a near-linear progression over time. Total retinal thickness increased approximately 1.56-fold within 270 minutes. Layer-specific swelling differed markedly, ranging from 1.5-fold in the outer nuclear layer to 2.2-fold in the inner plexiform layer. Despite these differences, swelling kinetics were remarkably conserved across retinal regions. Conclusions: Layer-specific OCT changes occur within minutes after retinal ischemia and exhibit highly reproducible, near-linear kinetics independent of retinal location. These findings establish a quantitative framework for time-resolved staging of retinal ischemia and may enable objective estimation of ischemia onset in clinical settings.