Associations of CAIDE Dementia Risk Score with MRI, PIB-PET measures, and cognition

Abstract

<h4><p>Background:</p><div><p>CAIDE Dementia Risk Score is the first validated tool for estimating dementia risk based on a midlife risk profile.</p><p>Objectives:</p></div><div><p>This observational study investigated longitudinal associations of CAIDE Dementia Risk Score with brain MRI, amyloid burden evaluated with PIB-PET, and detailed cognition measures.</p><p>Methods:</p></div><div><p>FINGER participants were at-risk elderly without dementia. CAIDE Risk Score was calculated using data from previous national surveys (mean age 52.4 years). In connection to baseline FINGER visit (on average 17.6 years later, mean age 70.1 years), 132 participants underwent MRI scans, and 48 underwent PIB-PET scans. All 1,260 participants were cognitively assessed (Neuropsychological Test Battery, NTB). Neuroimaging assessments included brain cortical thickness and volumes (Freesurfer 5.0.3), visually rated medial temporal atrophy (MTA), white matter lesions (WML), and amyloid accumulation.</p><p>Results:</p></div><div><p>Higher CAIDE Dementia Risk Score was related to more pronounced deep WML (OR 1.22, 95% CI 1.05–1.43), lower total gray matter (β-coefficient –0.29, <i>p</i> = 0.001) and hippocampal volume (β-coefficient –0.28, <i>p</i> = 0.003), lower cortical thickness (β-coefficient –0.19, <i>p</i> = 0.042), and poorer cognition (β-coefficients –0.31 for total NTB score, –0.25 for executive functioning, –0.33 for processing speed, and –0.20 for memory, all <i>p</i> < 0.001). Higher CAIDE Dementia Risk Score including APOE genotype was additionally related to more pronounced MTA (OR 1.15, 95% CI 1.00–1.30). No associations were found with periventricular WML or amyloid accumulation.</p><p>Conclusions:</p></div><div><p>The CAIDE Dementia Risk Score was related to indicators of cerebrovascular changes and neurodegeneration on MRI, and cognition. The lack of association with brain amyloid accumulation needs to be verified in studies with larger sample sizes.</p></div><p><br /></p><p><br /></p><p><br /></p> <p><br /></p><p><br /></p><p><br /></p><p><br /></p><p><br /></p><p><br /></p> <p><br /></p><p><br /></p><p><br /></p><p><br /></p><p><br /></p><p><br /></p><p><br /></p></h4>