Elders who complain that their memory ain’t what it used to be, and then measure up just fine on the standard battery of neurological tests, know something their doctors don’t, according to the results of a new imaging study. Their memory truly is declining, and for good reason—in such people, MRI measurements revealed brain atrophy comparable to that seen in subjects with mild cognitive impairment (MCI), a precursor to Alzheimer disease featuring measurable memory loss. The results, from Andrew Saykin and colleagues at Dartmouth Medical Center in Hanover, New Hampshire, show that for some people, subjective reports of memory problems can be an early indicator of neurodegeneration. The paper appeared in the September 12 issue of Neurology.

To figure out what was up with older people who consistently reported memory problems, the researchers performed MRI on 120 adults over 60 years old who had been extensively tested for memory impairment. Based on test results, information from self-reports and from informants, 40 of the group had been classified as completely normal (healthy control) and 40 were classified as mildly cognitively impaired (MCI). A group of 40 who tested normally but had more cognitive complaints than the healthy group were designated the cognitive complaint (CC) group. Though common factors causing memory complaints in older adults, like depression, medication, or white mater pathology, were ruled out in the CC group, they did have a lower composite score for verbal memory performance, albeit within the normal range. MRI scans were analyzed to measure hippocampal volume and grey matter density.

As expected, compared to healthy normal individuals the MCI group showed significantly lower gray matter in the medial temporal lobe and in distributed cortical regions, but surprisingly, the CC group also displayed a loss of gray matter density. Gray matter densities in the CC group were intermediate between the normal and MCI groups, suggesting that the CC subjects might represent a stage between normal aging and MCI, a clinical prodrome of AD (see ARF related news story).

Changes in gray matter density were more sensitive to the CC state than measurements of hippocampal volume made in the same subjects. In accordance with many studies, the MCI group showed significantly smaller hippocampal volume compared to healthy controls, while the CC group did not show a visible change in volume. Thus, density changes may be a better measure of the early preclinical stages of brain disease, and could be useful for early detection.

The changes in tissue density faithfully tracked with cognitive function: in the whole sample, lower verbal memory performance was associated with lower gray matter density in the left mediotemporal regions. Interestingly, higher levels of cognitive complaints correlated with decreased hippocampal gray matter density.

The results suggest that current neuropsychometric testing regimens are not sensitive enough to pick up subtle changes in mental function related to early brain atrophy, particularly among highly educated, high-functioning elders like those who made up the study group. In this regard, Saykin’s results recall a fascinating report last year on an aging chess player who complained of cognitive impairment because he could only plan four moves ahead in his game, where previously he could plan seven (Archer et al., 2005). Eventually, his prodigious mental capacities declined enough to be classified as MCI. When he died of unrelated causes shortly thereafter, an autopsy revealed severe AD.

That case, and this work, highlight the clinician’s difficulty in detecting subtle memory problems in people with high baseline functioning. Since the criteria for MCI are based on achieving a level of cognitive function relative to the population mean, people who start higher will necessarily have to decline more to reach criteria. The present work suggests that by taking measure of subjective complaints, and by following patients carefully over time, physicians may be able to catch pathology earlier.

“These findings highlight the importance of CCs in the clinical evaluation of older adults and suggest that those who present with significant CCs warrant evaluation and close monitoring over time,” Saykin and coauthors write. Taking note of patient-perceived, pre-MCI changes in mental function could open a door to early intervention with existing and future treatments.—Pat McCaffrey


  1. I think this is an important paper reporting on a well-conducted study.
    Although cross-sectional, the results imply a subtle continuum from
    cognitive complaints through to mild cognitive impairment and on to
    Alzheimer disease. Using a good-sized cohort of 40 subjects with
    cognitive complaints, 40 MCI subjects, and 40 controls, Saykin and
    colleagues used both voxel-based methods and region-of-interest outlining to show
    that relative to the control group, the MCI group had lower hippocampal
    volumes and lower grey matter densities in medial temporal lobe and
    elsewhere; importantly, they also showed that the group with cognitive
    complaints (but without cognitive deficits to meet MCI criteria) had
    hippocampal and grey matter volumes intermediate between the two groups.
    The paper therefore provides further, albeit circumstantial, evidence
    that structural changes—and in particular medial temporal lobe grey
    matter losses—may be detectable before cognitive deficits become
    manifest. It also reminds us that in certain settings, cognitive
    complaints should be taken seriously even when formal neuropsychometry
    appears within normal limits.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Brains in Transition: Signposts Point the Way from MCI to Alzheimer Disease

Paper Citations

  1. . Knight's move thinking? Mild cognitive impairment in a chess player. Neurocase. 2005 Feb;11(1):26-31. PubMed.

Further Reading

Primary Papers

  1. . Older adults with cognitive complaints show brain atrophy similar to that of amnestic MCI. Neurology. 2006 Sep 12;67(5):834-42. PubMed.