Premium
P1‐297: Associations between subjective memory impairment and hippocampal volume change in a prospective community‐based MRI study
Author(s) -
Stewart Robert,
Dufouil Carole,
Godin Ophelia,
Grassiot Blandine,
Crivello Fabrice,
Mazoyer Bernard,
Tzourio Christophe
Publication year - 2008
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 - 10.1016/j.jalz.2008.05.887
Subject(s) - dementia , odds ratio , verbal fluency test , neuroimaging , psychology , cognition , audiology , prospective cohort study , odds , medicine , cognitive decline , recall , neuropsychology , psychiatry , logistic regression , cognitive psychology , disease
Background: Subjective memory impairment (SMI) is one of few potential presenting symptoms for people with mild cognitive impairment. However, its relationship to underlying brain changes is controversial. We investigated the association between SMI and hippocampal volume changes in a prospective community-based neuroimaging study. Methods: In the Dijon centre of the 3C study, 1334 participants received an MRI scan at baseline and 4 year follow-up. SMI was ascertained at baseline and 2and 4-year follow-up. Associations were investigated between SMI at these time points and change in hippocampal volume. Covariates included age, sex, education, dementia incidence, and cognitive decline (MMSE, verbal fluency, visual memory). Results: SMI was present in 26.1%, 22.3% and 20.1% of the sample at the three respective time points. SMI at baseline did not predict subsequent hippocampal volume change, but SMI at 4-year follow-up was significantly associated with previous decrease in hippocampal volume (odds ratio across descending quintiles 1.15, 1.05-1.27). This was not altered after adjustment for age, sex, education and incident stroke (odds ratio 1.14, 1.03-1.25) or after further adjustment for current cognitive impairment or previous decline. It was strongest in participants who did not report SMI at baseline (odds ratio 1.22, 1.06-1.40). Conclusions: Subjective memory impairment was associated with underlying hippocampal volume loss, independent of measured cognitive function although more reflective of previous than subsequent change. SMI may therefore represent a realistic appraisal of brain changes before objective impairment can be detected.