Brain infarction and the clinical expression of Alzheimer disease. The Nun Study.
Level 3 - non-randomized controlled study
Prospective clinicopathological cohort study with post-mortem assessment
What was done
The study examined the relationship between brain infarcts and clinical dementia in 102 college-educated women aged 76 to 100 years from convents in the Midwestern, Eastern, and Southern United States (the Nun Study) who died and underwent autopsy. Cognitive function and dementia status were assessed with standard tests and clinical criteria during life. Post-mortem neuropathologic examination quantitated senile plaques and neurofibrillary tangles in the neocortex, identified lacunar and larger brain infarcts, and assessed circle of Willis atherosclerosis. Sixty-one participants met neuropathologic criteria for Alzheimer disease (AD) based on abundant senile plaques and some neurofibrillary tangles.
What was found
Among the 61 participants meeting neuropathologic criteria for AD, those with brain infarcts had poorer cognitive function and higher dementia prevalence than those without infarcts. Specifically, lacunar infarcts in the basal ganglia, thalamus, or deep white matter were associated with a marked increase in dementia prevalence compared to no infarcts (odds ratio [OR] 20.7, 95% CI 1.5-288.0). Participants with these subcortical lacunar infarcts developed dementia with fewer AD neuropathologic lesions than those without infarcts. Among the 41 participants without neuropathologic AD, brain infarcts were only weakly associated with poor cognitive function and dementia. Across all 102 participants, circle of Willis atherosclerosis was strongly associated with both lacunar and large brain infarcts.
Why it matters
This paper demonstrates that cerebrovascular disease, particularly subcortical lacunar infarction, lowers the threshold of Alzheimer neuropathology required for clinical dementia to emerge.
Limits
The study population was restricted to elderly, college-educated women living in convents, limiting generalizability to men and more diverse socioeconomic or lifestyle groups. The sample size was modest (n=102 total, 61 with AD pathology), leading to an extremely wide confidence interval for the primary odds ratio (1.5-288.0). The abstract does not provide exact infarct volume measurements or the precise temporal sequence of infarct onset relative to cognitive decline.
Cited by
- supports In the Nun Study, some participants had hallmark Alzheimer's disease pathology including amyloid plaques and neurofibrillary tangles upon post-mortem examination, but remained cognitively high functioning during life.