Cerebral small vessel disease encompasses a myriad of pathologies, often with heterogenous underlying etiologies and pathophysiological mechanisms, that affect different elements of the cerebral vasculature system. Arteriolosclerosis is a common small vessel pathology affecting the deep penetrating small arteries and arterioles, involving subcortical brain regions, both white matter and grey matter nuclei. It is characterized by progressive thickening, hardening, and fibrotic changes of the vessel wall combined with the constriction of the vessel lumen. Cellular changes include damage to the endothelial cell layer, intimal hyperplasia, loss of smooth muscle cells, and fibrohyalinotic thickening of the vessel wall, resulting in loss of elasticity within vessels (Blevins et al., 2021, Gutierrez et al., 2024).

Clinicopathological studies, including work from our group, have shown arteriolosclerosis pathology evaluated at the time of autopsy is associated with cognitive impairment and higher odds of Alzheimer’s dementia, even after accounting for other potential neurodegenerative and cerebrovascular pathologic confounders in the brain (Arvanitakis et al., 2016, Kryscio et al., 2016, Snyder et al., 2015). More recently, we and others have demonstrated that arteriolosclerosis pathology is associated with neurodegenerative pathologies that drive subsequent brain atrophy, specifically phosphorylated tau tangles and cytoplasmic aggregation of TAR DNA-binding protein 43 (TDP-43) (Kapasi et al., 2021, Kapasi et al., 2022, Nho and Saykin, 2016); supporting the notion that a complex relationship exists between small vessel disease and neurodegeneration. However, the relationship between arteriolosclerosis and magnetic resonance imaging (MRI) brain markers of neurodegeneration is poorly understood.

Because arteriolosclerosis pathology can only be formally diagnosed on postmortem tissue, our group recently developed a novel, automated, freely available in-vivo MRI-based classifier for ARTerioloSclerosis, named ARTS (Makkinejad et al., 2021). ARTS takes three modalities of raw in-vivo MRI data as input, conducts all necessary processing automatically and outputs a score that represents the likelihood a person suffers from arteriolosclerosis, where higher ARTS score represents higher likelihood of arteriolosclerosis. In this study, we utilized the ARTS classifier and cortical thickness measures obtained from the same MRI to test the hypothesis that higher ARTS score is associated with lower cortical thickness and further examined whether cortical thickness mediated the relationship between ARTS and cognition.