Study design: NYU ADRC participant cohort from 2010-2019. Clinical consensus diagnosis of SCD: NACC UDS 2 & 3, NYU specific measures, Global Deterioration Scale = 2. Retrospective crosssectional analysis of cognitive testing and affective symptoms (N = 107). Measures: SCD: Brief Cognitive Rating Scale (BCRS), Cognitive Change Index (CCI), Affective: Beck Anxiety Inventory (BAI), Beck Depression Inventory (BDI), Neuropsychiatric Inventory Questionnaire (NPI-Q), Cognitive: Montreal Cognitive Assessment (MoCA), Mini Mental State Examination (MMSE), NACC UDS subtests: Trail Making Test (A&B), and Verbal fluency Clinical comorbidities (N = 46), ApoE status (N = 77), Structural MRI (N = 17).

Statistical analyses: Unpaired t-test for magnitude comparisons, ANOVA for individual responses. Simple and multiple linear regression models when appropriate, with models adjusted for demographic and clinical variables that were statistically significant in the first analysis.

Results:

Demographic and vascular factors were primarily associated with anxiety. Hispanic ethnicity associated with both anxiety and depression. Depressive symptoms had a stronger correlation with subjective cognitive complaints. CCI and BCRS items significantly associated with higher depression scores. Specific domains associated with both anxiety and depression: psychomotor function, praxis and calculation. Overall, cognitive performance measures showed limited and inconsistent associations with affective symptoms after controlling demographics. Lower global cognitive performance (MMSE) was associated with higher anxiety in unadjusted models. No significant associations were observed between MMSE or MoCA scores and depression. Executive function, as measured by TMT-A, was associated with higher anxiety in unadjusted models, but not after adjustment. Depression was associated with poorer performance on verbal fluency (vegetables) in unadjusted analyses. Greater anxiety and depression were associated with reduced cortical thickness in frontotemporal regions. Specifically, thinner cortex in the right rostral middle frontal gyrus (RMFG) was associated with both higher anxiety and depression across all models. Depression was additionally associated with reduced thickness in the right pars orbitalis. Anxiety was associated with reduced thickness in the right medial orbitofrontal cortex, left caudal anterior cingulate, and left parahippocampal gyrus. These findings remained consistent after adjustment for age and sex. 

Conclusions: 

Our findings may suggest that although overlapping, anxiety and depression in SCD may reflect partially different clinical and neurobiological processes. Hispanic ethnicity was associated with both anxiety and depression, with a stronger effect observed for anxiety. Depression is more strongly associated with subjective cognitive complaints. Anxiety is associated with frontotemporal cortical thinning and vascular risk factors.

Future directions:

Longitudinal studies are needed to determine whether the distinct cognitive and neuroanatomical profiles associated with anxiety and depression in individuals with SCD predict differential risk for cognitive decline or dementia. Further research should investigate the mechanisms underlying these distinct associations and evaluate their potential as early markers of disease progression.

References: 

1. Jessen F et al. Alzheimers Dement. 2014;10(6):844-852.
2. Botto R et al. Neurol Sci. 2022;43(7):4107-4124.
3. Perin S et al. J Affect Disord. 2022;296:559-566.
4. Desai R et al. Ageing Res Rev. 2021 Nov;71:101419.

Acknowledgements:

This study was funded by NIH P30AG066512 and the Louis J and June E Kay Foundation.