Abstract Number: Esoc2026a66 Retinal and Choroid Alterations in Carotid Dissection vs. Atherosclerosis: Insights Into Divergent Ischemic Mechanisms

Abstract Background and aims This study aimed to comprehensively characterize retinal and choroidal alterations in carotid artery dissection (CAD), compare them with those in atherosclerotic carotid artery stenosis (CAS), and analyze their associations with clinical and neuroimaging features, including cerebral infarction, stenotic degree, and infarction location. Methods This observational study included CAD patients and stenotic-degree-matched CAS patients. Retinal ischemic signs—cotton wool spots (CWS), retinal artery occlusion (RAO), and paracentral acute middle maculopathy (PAMM)—were evaluated using optical coherence tomography (OCT) and angiography (OCTA) centered at foveal. Retinal vascular metrics (superficial and deep vascular complex density) and choroidal metrics (choriocapillaris density, choroidal thickness, choroidal vascular volume CVV, and the choroidal vascular index CVI) were analyzed. Results The final analysis included 863 eyes from 467 participants. CAD patients showed a significantly higher prevalence of PAMM than CAS patients (6.4% vs. 0.61%, p 0.001). Eyes ipsilateral to CAD exhibited reduced retinal vascular density compared with both contralateral and control eyes. Furthermore, CAD-affected eyes demonstrated significantly higher choriocapillaris density, CVV, CVI, and choroidal thickness relative to both control eyes and CAS-affected eyes. PAMM presence was significantly associated with cerebral infarction and severe stenosis/occlusion. Subcortical infarctions were more frequently associated with PAMM, while cortical infarctions tended to co-occur with CWS and RAO. Conclusions CAD exhibit more pronounced retinal hemodynamic impairment, higher thromboembolic risk, and distinct vascular alteration patterns compared with atherosclerotic stenosis. Retinal imaging provides a non-invasive window into CAD pathophysiology, offering valuable insights for clinical assessment and risk stratification. Conflict of interest Le Cao, Hang Wang, and Bo Wu: nothing to disclose