Premium
Depressive symptom profiles predict dementia onset and longitudinal changes in hippocampal and white matter lesion volume in older persons: The AGES‐Reykjavik study
Author(s) -
Gerritsen Lotte,
Geerlings Mirjam I,
Sigurdsson Sigurdur,
Jonsson Palmi V,
Gudnason Vilmundur,
Launer Lenore J
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.042422
Subject(s) - apathy , dementia , late life depression , depression (economics) , medicine , geriatric depression scale , hazard ratio , population , hyperintensity , psychiatry , gerontology , disease , confidence interval , depressive symptoms , cognition , magnetic resonance imaging , hippocampal formation , environmental health , radiology , economics , macroeconomics
Background Late‐life depression (LLD) is a major risk factor for dementia, including Alzheimer’s disease. However, not every patient with LLD will develop dementia or show neurodegeneration. Since LLD is heterogeneous in symptom presentation subtypes may exist within LLD that have a higher risk of dementia than others. We aimed to 1) identify subtypes in LLD using latent class analyses (LCA), and 2) examine how these subtypes were related to dementia and MRI markers of neurodegenerative and vascular pathology over time. Method For 2430 participants (75±6 years; 41% male) without dementia from the population‐based AGES‐Reykjavik Study the Geriatric Depression Scale (GDS‐15) was administered and 1.5T brain MRI was performed at baseline and follow‐up after 5 years. Dementia, including AD, was assessed according to international criteria during an average follow‐up of 9 years. We performed LCA with GDS‐15 items examining 1 to 7 classes. Cox regression analyses adjusted for age, sex, and education were used to estimate the Hazard Ratios of LLD subtype with incident dementia. ANCOVA was used to examine the difference in hippocampal and white matter lesion volume (WML) from baseline to follow‐up, adjusted for age, sex, education and ICV. Result LCA with 5 classes rendered best fitting results: no depression (57%); apathy (31%); apathy with emptiness (2%), mild depression (8%) and severe depression (2%). Severe depression (HR=1.87; 95%CI: 1.2‐2.9), mild depression (HR=1.28; 95% CI:1.1‐1.6) and apathy (HR=1.39; 95% CI:1.2‐1.6) had increased risk of developing dementia compared to the no depression group (Figure 1). Figure 2 shows the hippocampal decline(Figure 2A) and increase in WML(Figure 2B) per LLD subgroup; compared to the no depression group the subtype with severe LLD showed most hippocampal decline (F(4,2429)= 2,28; p =0.04) and increase in WML (F(4,2429)= 2,72; p =0.02). Conclusion This study shows that different subtypes of LLD can be identified. Severe depression as well as apathy were associated with increased risk of dementia, but only severe LLD was related to progression of markers of neurodegeneration and vascular pathology.