9 July 2006. Baby boomers joke about forgetting, usually without thinking twice, but are such minor memory lapses just a normal part of the aging process, or could they be a sign of worse things to come? Though people with mild cognitive impairment (MCI) can progress to full-blown dementia, subtle changes in episodic memory have not generally been considered a harbinger of Alzheimer disease (AD)—until now. Dave Bennett and colleagues at Rush Alzheimer's Disease Center, Chicago, report in the June 27 Neurology that after excluding identifiable MCI cases from postmortem evaluation, a significant number of cognitively normal older persons met neuropathological criteria for AD on death. Those individuals also scored lower on tests of episodic memory.
It is well established that older persons may harbor AD pathology—neuritic plaques, diffuse plaques, and neurofibrillary tangles—yet exhibit no signs of dementia. However, in past studies, some of those “normal” subjects turned out, in retrospect, to have had MCI, which raises the question: can AD pathology ever be found in someone without MCI, or AD? Bennett and colleagues address this by examining postmortem brain samples from 134 participants from the Religious Order Study (98 persons) and the Rush Memory and Aging Project (36 persons). Each participant had been evaluated annually for cognition using 19 cognitive performance tests, including the Mini-Mental State Examination (MMSE), episodic memory, working memory, and visuo-spatial ability. Bennett and colleagues diagnosed on death AD based on the National Institute on Aging-Reagan (NIA-Reagan) criteria. These combine guidelines set by the Consortium to Establish a Registry for AD (CERAD) with Braak stages and determine the probability that clinical dementia has been caused by AD pathological lesions.
Autopsies revealed neuritic plaques and neurofibrillary tangles (NFTs) in several cortical regions and the hippocampus. Braak stages were assigned based on the distribution and severity of NFT pathology. Additionally, neuropathologic diagnoses (low, intermediate, or high likelihood of AD) were made based on estimates of neuritic plaque density. One-third of the participants met NIA-Reagan criteria for intermediate likelihood of AD, and two people (1.5 percent) met the criteria for high likelihood of AD. Importantly, both groups (high and intermediate) exhibited subtle deficits in episodic memory, suggesting that these slight impairments, thought to represent consequences of normal aging, might actually indicate the presence of AD pathology. In fact, linear regression models adjusted for age, sex, and education demonstrated that these two groups scored roughly a quarter standard unit lower on episodic memory tests—the only cognitive performance tests in which there were any significant differences among the persons tested.
Bennett and colleagues discuss the possibility that some sort of neural reserve allows some older persons to tolerate considerable amounts of AD pathology without showing symptoms. The redundancy found in the functional organization of the brain may result in some sort of compensation by alternative neural networks. Such a compensatory mechanism, coupled with environmental factors such as education, could potentially raise the threshold for tolerance of AD neuropathology, delaying cognitive manifestations of the disease.
In an editorial in the same issue of Neurology, Carol Lippa of Drexel University College of Medicine in Philadelphia and John Morris, Washington University School of Medicine in St. Louis, suggest that “even before a person meets formal criteria for MCI, the toxic cascades that eventually result in overt dementia are already in play.” In fact, the early decline in episodic memory observed in the study may represent a “pre-MCI state that now should be the target of research efforts in the early detection of AD,” they write.
Still, these careful characterizations of neuropathology in normal aged persons are not flawless. In fact, Bennett and colleagues point out some potential limitations of the study, including the brevity of the interviews used to assess cognition and the lack of neuroimaging. Additionally, Lippa and Morris question the reliability of self-reports of cognitive function, and highlight the need for retrospective interviews with family members about cognitive function during the last interval before death. They also noted the emphasis on neuritic plaques and tangles as diagnostic criteria, and the lack of data on diffuse plaques, which are typically associated with early AD. But despite these limitations, Bennett and colleagues may have identified the earliest point in a process that eventually leads to AD, a stage during which treatment may be most effective.—Erene Mina.
Bennett DA, Schneider JA, Arvanitakis Z, Kelly JF, Aggarwal NT, Shah RC, Wilson RS. Neuropathology of older persons without cognitive impairment from two community-based studies. Neurology. June 27, 2006;66:1837-1844. Abstract