Cerebral infarction, as the result of atheroma in large blood vessels, was one of the first explanations for dementia in older people. However, this was a relatively uncommon finding in those presenting with dementia and it was soon realized that subcortical infarctions due to non-atheromatous changes in small blood vessel walls were more important factors in cognitive decline. Furthermore, small vessel disease (SVD) producing lacunar infarcts (LI) and white matter changes increased dementia severity in individuals with only moderate Alzheimer's disease (AD) pathology. Magnetic resonance imaging (MRI) found subcortical (lacunar) ‘silent infarcts’ and white matter changes were surrogate markers for SVD and increased significantly the likelihood of developing dementia. Detailed histopathological studies of SVD showed it was in fact a range of pathologies, but the aetiology of the most common type (arteriosclerosis/lipohyalinosis) is difficult to explain. Theories to explain SVD include effects of hypertension, a leaky blood–brain barrier (BBB), amyloid accumulation, microbleeds, venous changes and endothelial inflammation. Endothelial expressed inflammatory molecules could represent potential biomarkers of SVD. Many of the risk factors associated with SVD are also present in AD, such as diabetes, systemic hypertension and ageing; all of these can target the endothelium with the potential to initiate SVD, resulting in ischaemic changes and neurodegeneration, including AD pathology.

Subcortical ischemic vascular disease and dementia (SIVD) incorporate small vessel disease as the chief vascular etiology, lacunar infarct and ischemic white-matter lesions (WMLs) as primary type of brain lesions, subcortical location as the primary location of lesions, and subcortical syndrome as the primary clinical manifestation. It incorporates two clinical entities: Binswanger's syndrome and the lacunar state. Patients with SIVD present with extensive white-matter lesions and multiple lacunae on neuroimaging. SIVD is expected to be a more homogenous subtype of vascular cognitive impairment and dementia. Recently modified NINDS-AIREN research criteria for SIVD have been proposed. Further empirical research is needed to refine the syndrome and stages and validate the brain imaging criteria, as well as to detail the natural history and outcomes of SIVD.