Introduction: Carotid near-occlusion is a severe carotid stenosis causing distal artery collapse of varying degree. Near-occlusion is often divided into a “full collapse” group with a threadlike distal lumen, and the often overlooked “without full collapse” group with a normal-appearing, albeit small, distal lumen. By this division by appearance, symptomatic near-occlusion with full collapse has been reported to have worse short-term prognosis than those without full collapse, no other division has been assessed for prognosis. The aim of this study was to assess if a measurement based definition of full collapse might improve prognostic discrimination. Methods: 99 consecutive patients with symptomatic near-occlusion diagnosed on CT-angiography were included. The risk of preoperative recurrent ipsilateral ischemic stroke within 28 days of presenting event was assessed with Kaplan-Meier analysis, censoring at revascularization. We assessed residual stenosis diameter, distal ICA diameter, ICA-ratio (side-to-side), and ICA-ECA ratio as risk markers. Results: By appearance, the 28-day risk of stroke tended to be higher for full collapse (27%, 11/42) than without full collapse (11%, 6/57), p=0.054 (figure). The best new definition of full collapse was distal ICA diameter ≤2.0 mm and/or ICA ratio ≤0.47. 10 patients were reclassified by this new definition compared to appearance definition, 5 in each direction. By the new definition, 28-day risk of stroke was higher in full collapse (34%, 14/42) than without full collapse (5%, 3/57), p<0.001 (fig 1B). Conclusions: Compared to the appearance based definition, our new definition of separating near-occlusions into with and without full collapse yields similar groups sizes but better prognostic discrimination. This new definition could be used as inclusion criteria in future treatment trials.