Carotid arteries in cerebral small vessel disease and dementia

Carotid artery disease (CAD) is a recognised cause of stroke. However, the relationships between CAD, cerebral small vessel disease (SVD) and dementia remain unclear. We hypothesised that CAD in older individuals contributes to cerebral SVD pathology by altering cerebral perfusion. We performed a clinicopathological study in patients from the Cognitive Function After Stroke (CogFAST) study and prospectively recruited patients with various dementia diagnoses and evidence of cerebral SVD. In addition to brain tissues, we collected postmortem samples of the internal carotid arteries (ICA) from these cohorts in the Newcastle Brain Tissue Resource. Standard neuropathological examination was performed for diagnosis and assignment of the cases per current diagnostic criteria for vascular and neurodegenerative dementias, which were assessed for the presence of vascular pathology including the degree of stenosis and sclerosis in vascular tissues. We evaluated a total of 159 ICA samples and brain tissues from all cases with evidence of SVD. Severity of ICA stenosis and sclerotic index correlated strongly with both clinical stroke and brain infarction (P < 0.001). More than 90% of the subjects had one subtype of ICA lesion in the order: intimal thickening > fibrocalcific > fibrous cap (thick) > fibrous cap (thin) > thrombus group with a strong inflammatory reaction in fibrocalcific atheromas. Regression analyses showed that ICA stenosis was positively correlated to both SVD pathology scores (P < 0.034) and the total number of vascular lesions (P < 0.001). ICA stenosis was also related to dementia caused by cerebrovascular disease (P < 0.001) and by mixed pathologies characterised by Alzheimer's disease and SVD (P = 0.025). Severity of stenosis was related to subcortical and white matter (WM) vascular lesions within the anterior circulation. ICA stenosis and sclerosis were moreover correlated with the total intracranial artery pathology scores (P < 0.001). In the CogFAST group analysis, we observed that MMSE and CAMCOG scores were lower in subjects with moderate to severe stenosis scores compared to the mild stenosis group (P < 0.05). In these CogFAST cases, by far the majority of lesions in the WM were small in size (< 5 mm, range 72-91%) but not in the cortex or basal ganglia and thalamus. Linear regression analysis further indicated that there were greater numbers of these small lesions in the WM with increasing severity of ICA stenosis (P < 0.05). Our observations suggest carotid atherosclerosis promotes cerebral SVD types of change within the intracerebral arteries. It is conceivable that extracranial ICA pathology may influence perfusion and integrity of subcortical structures including the deep WM.