The predictive value of cerebral ischemic lesions for dementia

• Background & Objective Changes in white matter are known to be related to dementia and cognitive decline. We examined whether CT-detected white matter lesions (WMLs) predicted dementia or cognitive decline in a population followed for five years. Methods A representative sample of persons aged 70-82 years were examined with a neuropsychiatric examination and a CT scan in 2000-2001 (N=657). A follow-up examination was performed in 2005 (N=503). People without dementia at the time of CT scanning and who participated in the examinations in 2005 were included (N=497). The Wahlund scale was used to define WMLs using a score from 0 (no lesions) to 3 (severe lesions). Cognitive status was assessed with the Mini Mental State Examination (MMSE). The difference in MMSE score from 2000 to 2005 (∆MMSE) was used as a general measure on cognitive decline. Differences in performance on other neuropsychiatric tests were included to get more sensitive measures of change in cognitive function. Regression analyses were applied to analyze whether WMLs could predict the outcome of dementia or cognitive decline five years later. Results Thirty-seven people were diagnosed with dementia in 2005 (7.4%), of these 12 (32%) had any WML of severity >1 on the CT performed five years earlier. For the non-demented group the comparable number was 77 (17%). Significant cognitive decline according to ∆MMSE was seen in the group who were diagnosed with dementia. Preliminary regression analysis shows that cognitive decline and parieto-occipital WMLs predicted onset of dementia. Conclusion The findings suggest that WMLs are weak predictors of dementia. Only WMLs in parieto-occipital areas were related to dementia five years later. Further analyses will show the relation between WMLs and decline in cognitive symptoms.

Subject headings

MEDICIN OCH HÄLSOVETENSKAP  -- Klinisk medicin -- Psykiatri (hsv//swe)

MEDICAL AND HEALTH SCIENCES  -- Clinical Medicine -- Psychiatry (hsv//eng)

Publication and Content Type

vet (subject category)

kon (subject category)