Abstract
Cortical microinfarcts (CMIs) observed in brains of patients with Alzheimer's disease tend to be located close to vessels afflicted with cerebral amyloid angiopathy (CAA). CMIs in Alzheimer's disease are preferentially distributed in the arterial borderzone, an area most vulnerable to hypoperfusion. However, the causal association between CAA and CMIs remains to be elucidated. This study consists of two parts: (1) an observational study using postmortem human brains (n = 31) to determine the association between CAA and CMIs, and (2) an experimental study to determine whether hypoperfusion worsens CAA and induces CMIs in a CAA mouse model. In postmortem human brains, the density of CMIs was 0.113/cm(2) in mild, 0.584/cm(2) in moderate, and 4.370/cm(2) in severe CAA groups with a positive linear correlation (r = 0.6736, p < 0.0001). Multivariate analysis revealed that, among seven variables (age, disease, senile plaques, neurofibrillary tangles, CAA, atherosclerosis and white matter damage), only the severity of CAA was a significant multivariate predictor of CMIs (p = 0.0022). Consistent with the data from human brains, CAA model mice following chronic cerebral hypoperfusion due to bilateral common carotid artery stenosis induced with 0.18-mm diameter microcoils showed accelerated deposition of leptomeningeal amyloid β (Aβ) with a subset of them developing microinfarcts. In contrast, the CAA mice without hypoperfusion exhibited very few leptomeningeal Aβ depositions and no microinfarcts by 32 weeks of age. Following 12 weeks of hypoperfusion, cerebral blood flow decreased by 26% in CAA mice and by 15% in wild-type mice, suggesting impaired microvascular function due to perivascular Aβ accumulation after hypoperfusion. Our results suggest that cerebral hypoperfusion accelerates CAA, and thus promotes CMIs.