Cerebral small vessel disease (SVD) accounts for 20% of ischaemic stroke and is the major cause of vascular dementia. Disease of the small perforating arteries supplying white matter and the deep grey matter nuclei results in small focal areas of complete infarction (lacunar infarction) and/or areas of diffuse incomplete infarction (leukoaraiosis) which neuropathologically show ischaemic demyelination, axonal loss and gliosis. The pathogenesis is incompletely understood. It has been suggested there may be two pathological types of cerebral SVD.(1) Microatheroma at the origins or in the proximal perforating arteries resulting in single larger lacunar infarcts without leukoaraiosis, and a diffuse small vessel arteriopathy affecting the smaller perforating arteries resulting in multiple smaller lacunar infarcts and leukoaraiosis. The latter subtype has been referred to as ischaemic leukoaraiosis.(2) In this subtype hypertension is the major risk factor, and impaired autoregulation and reduced cerebral blood flow appear important, resulting in leukoaraiosis, initially in those internal watershed areas in which arterial perfusion pressure is lowest. Both cerebral blood flow and cerebral autoregulation are dependant on normal endothelial function and nitric oxide release.(3) A number of lines of evidence implicate impaired endothelial function in cerebral SVD pathogenesis including pathological data, circulating endothelial markers, and imaging findings.(2) Genetic factors appear important, with the heritability of MRI determined cerebral SVD in community populations estimated to be as high as 60%.(4) Candidate gene association studies have implicated genes affecting endothelial function including eNOS and components of the renin-angiotensin system. However, larger studies have been unable to consistently confirm these findings.(5) Almost all candidate gene studies have been underpowered to date. Larger multicentre collections are now being performed to evaluate the responsible genetic variants.