The response to ischemic injury in the brain is different to the response to ischemic injury in other organs and tissues. Almost exclusive to the brain, and for unknown reasons, dead tissue liquefies in response to ischemia by the process of liquefactive necrosis. However, the data we present here indicate that at the macroscopic, microscopic, and molecular level, liquefactive necrosis strongly resembles atherosclerosis. We show that chronic stroke infarcts contain foamy macrophages, cholesterol crystals, high levels of osteopontin and matrix metalloproteinases, and a similar cytokine profile to atherosclerosis. Excessive cholesterol loading of macrophages is a principal driver of atherosclerosis. Therefore, because cholesterol is an important structural component of myelin, liquefactive necrosis in response to stroke may be caused by an inflammatory response to myelin debris that is prolonged by the formation of cholesterol crystals within macrophages. We propose that this results in the chronic production of high levels of proteases, which in a partially osteopontin dependent mechanism, causes secondary neurodegeneration and encephalomalacia of the surrounding tissue. In support of this, we show that genetically ablating osteopontin substantially reduces the production of degradative enzymes following stroke, reduces secondary neurodegeneration, and improves recovery.These findings suggest that treatments that prevent atherosclerosis or target the regression of All rights reserved. No reuse allowed without permission.