Segmentation and morphometry of intracranial internal carotid artery calcification in relation to brain atrophy

Abstract

Intracranial internal carotid artery calcification (iICAC) is a form of intracranial arteriosclerosis and is associated with an elevated risk of stroke and dementia. However, iICAC’s relationship with brain atrophy remains poorly understood. We aimed to automatically quantify iICAC morphometric characteristics and evaluate their associations with regional brain volumes (BVs). We developed an automated approach to compute iICAC surface area and thickness from CT brain scans in a sample of physically active South American subsistence farmers (n = 1,232, age range: 40 years to 92 years, 48.1% female, 794 Tsimane and 438 Moseten). Linear regression models were used to assess associations between two iICAC features and regional BVs, adjusted for age, sex, population, and total intracranial volume. Significant negative relationships were found between regional BVs and iICAC surface area, but not iICAC thickness. Frontal, parietal, temporal, and subcortical BVs exhibited significant negative associations with iICAC surface area (standardized $$\beta$$ range: -0.146 to -0.066, p ≤ 0.013), while the occipital BV did not (standardized $${\beta}_{left}$$ = -0.035, p = 0.249; $${\beta}_{right}$$ = 0.007, p = 0.810). Subcortical BVs demonstrated the strongest negative associations with iICAC surface area (standardized $${\beta}_{left}$$ = -0.146, p < 0.001; $${\beta}_{right}$$ = -0.139, p < 0.001). iICAC surface area—assumed to reflect arterial stiffness—shows a stronger relationship with regional BV loss than iICAC thickness—assumed to indicate arterial stenosis. The findings suggest that brain regions primarily supplied by the anterior circulation are more vulnerable to iICAC-related atrophy. Subcortical BVs showed the strongest negative associations with iICAC surface area, with region-specific analyses identifying significant effects in the putamen, thalamus, hippocampus, amygdala, pallidum, and ventral diencephalon, suggesting heightened vulnerability of deep gray-matter structures to iICAC-related atrophy.

Description

Citation