| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From Beth Israel Deaconess Medical Center (M.H.S., C.B., T.T.P.) and Massachusetts General Hospital (M.N., T.H.-W.), Boston, Massachusetts.
Address reprint requests to: Margery H. Silver, Harvard Medical School Division on Aging/Beth Israel Deaconess Medical Center, One Deaconess, Road, CC105, Boston, MA 02215. Email: margery_silver{at}hms.harvard.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONDefining Disease-Free Aging in...METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
METHODS: Fourteen of 74 centenarians who underwent annual extensive neuropsychological evaluation proceeded to postmortem neuropathological examination. CERAD criteria, emphasizing neuritic amyloid plaques and Braak and Braak staging of neurofibrillary tangles were used to assess the 14 brains.
RESULTS: Neuropsychological and neuropathological findings correlated well for four subjects with no dementia on testing (CDR = 0) and for six subjects with CDR scores in the dementia range (CDR = 1–5). In the latter group, Alzheimer’s disease was diagnosed in four brains; Pick’s disease was an etiological factor in the fifth and hippocampal sclerosis in the sixth. Correlation was low for four subjects: two subjects with no dementia on neuropsychological testing met CERAD neuropathological criteria for possible AD; two subjects with dementia on testing did not meet CERAD criteria for definite Alzheimer’s disease and had otherwise minimal changes to correlate with the cognitive findings.
CONCLUSIONS: Lack of correlation between level of cognitive functioning and brain pathology in two subjects with no dementia raised the question of whether a functional reserve delayed the functional expression of pathological changes. For two subjects with dementia on testing, there appeared to be no sufficient pathological explanation for the extent of the cognitive changes; depression and such factors as environment, sensory impairment, and medical illness may all have played a role. There may also have been neuropathologic changes not detected by current methods.
Key Words: Alzheimer’s, • centenarians, • cognition, • dementia, • neuropathology, • neuropsychology.
Abbreviations: AD = Alzheimer’s disease;; CDR = clinical dementia rating;; CERAD = Consortium to Establish a Registry for Alzheimer’s Disease;; DSM III-R = Diagnostic and Statistical Manual of Mental Disorders, Third Edition-Revised;; GDS = Geriatric Depression Scale;; MDRS = Mattis Dementia Rating Scale;; MMSE = Mini-Mental State Examination;; NECS = New England Centenarian Study;; NFT = neurofibrillary tangles;; NIA = National Institute of Aging;; TIA = transient ischemic attack;; WAIS = Wechsler Adult Intelligence Scale.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONDefining Disease-Free Aging in...METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Although there has been concern about the burgeoning of this extreme old population and its effects on health and social systems, little is known about level of cognitive functioning and prevalence of dementia and other neurocognitive diseases in this group. Most research on cognition in aging has been conducted on persons aged 60 to 80 years. In fact, even those over 60 were rarely studied until the 1970s. As centenarian studies progress, it is becoming apparent that even assumptions based on findings about younger old groups are not necessarily true for 100-year-olds. For example, contrary to estimates in the past that all centenarians would be demented (8), prevalence of centenarians with no dementia in a New England population was found to be 21% (4).
|
Defining Disease-Free Aging in Centenarians |
|---|
|
TOPABSTRACTINTRODUCTIONDefining Disease-Free Aging in...METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Neurocognitive functioning in centenarians has been examined from a number of different perspectives and with a variety of methods. Cross-sectional studies have compared nondemented centenarians with younger cohorts to define the usual pattern of cognitive abilities in disease-free 100-year-olds. Poon and colleagues (11) found that centenarians performed significantly lower on verbal and performance measures than 60- to 80-year-olds but did not differ in their ability to solve practical problems. Samuelsson and colleagues (12) reported similar findings with Swedish centenarians on new learning and working memory tests compared with 16- to 57-year-olds.
Other studies have focused on presence rather than absence of disease in centenarians, with the goal of determining prevalence of dementia in this age group. These studies have been difficult to compare because of varied methodologies and different criteria for a diagnosis of dementia. Thomassen and colleagues (8), in a Dutch study that used participant or informant questionnaires, found 89% of centenarian subjects to be moderately to severely demented. Heeren and colleagues (13) administered the MMSE and the Geriatric Mental State Schedule to estimate the prevalence of dementia in three groups of oldest old in a Dutch community. They found that 50% of the 95+ population had dementia. Hagberg and fellow researchers (14), comparing prevalence of dementia in Japanese and Swedish centenarians, reported prevalence of dementia between 40% and 63% and suggested that the wide range might be due to differences in sampling and criteria for dementia.
From a different perspective, another group of studies has focused on neuropathological examination of centenarian brains. These studies have presented difficulties because neuropsychological criteria for dementia, such as the CERAD protocol and Braak and Braak staging (described in detail in the Methods section), have been developed in younger populations (15). In fact, Gold and colleagues (16) have suggested, based on data from a comparison of Clinical Dementia Rating (CDR) scores and Braak staging from 116 nonagenarian and centenarian brains, that the neurofibrillary tangle staging described by Braak and Braak (17) be revised for rating of AD in the oldest old. They found that stages I and II, which are clinically silent in younger adults, are not necessarily clinically silent in the oldest old, and although the highest stage, stage VI, is usually associated with dementia in younger adults, stage IV is consistently associated with at least mild dementia in the oldest old.
Moreover, neuropathological studies have been difficult to compare because of lack of uniformity in the methods of determining a premorbid clinical diagnosis of dementia. For example, Mizutani and Shimada (18) found no cases of AD in neuropathological evaluation of 27 Japanese centenarians, but no clinical dementia evaluation had been performed premorbidly. It was reported only that the subjects were in "good mental condition" at the time of their deaths. Samuelsson’s group (12), using informant and participant interviews, found "Alzheimer–like changes. . . short of Alzheimer’s disease" in eight centenarian brains. A French study of 12 centenarians, using DSM-III standards (19), reported that only three were rated demented and of the three, only one met pathological criteria for AD (20).
Previous work by Silver and colleagues (21) reported on both dementia prevalence and neuropathological findings. The full range of neurocognitive functioning was measured using a comprehensive testing battery, and dementia prevalence was documented for centenarians in eight towns in the area surrounding Boston. This study suggested that, contrary to assumptions made by others based on research of younger subjects, dementia is not inevitable in centenarians (8).
To identify neurodegenerative diseases accurately, particularly AD, and other diseases that cause dementia and to examine correlations between neuropsychological evaluations and neuropathological findings, the brains of a subgroup of centenarians who had consented to autopsy and died after neuropsychological exam were studied. In a group of six brains, there were no cases of AD (4). The correlation results raised the question of how to define disease-free aging. In some cases, clinical and neuropathological conclusions did not correlate. Two centenarians with dementia on neuropsychological testing did not meet criteria for probable AD or other pathological entities. In regard to individuals who were not demented on neuropsychological examination but have pathological markers of AD, it has been suggested that a functional brain or cognitive reserve may prevent or delay the expression of neuropathological changes (22, 23).
The present study expands on the work reported previously. It addresses questions about disease-free aging in centenarians, reporting on the ongoing work conducted by the NECS in eight towns in the Boston area and focusing on the correlation between neuropsychological and neuropathological findings. At this point in time, an additional eight neuropsychologically tested subjects who consented to autopsy have died. In this article, we report on the correlation of neuropsychological testing with neuropathological findings in 14 centenarian brains evaluated primarily for evidence of Alzheimer changes, namely neuritic amyloid plaques and neurofibrillary tangles. The larger sample provides us with a greater range of neuropsychological and neuropathological findings, providing additional insight into how to define disease-free aging as well as an opportunity to discover whether certain findings—eg, no AD in the six earlier cases—were an artifact of sample size.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONDefining Disease-Free Aging in...METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
The sample was composed of 86% women and 14% men. Their ages ranged from 100 to 110 years. Nineteen (26%) of the centenarians lived with their family; 5 (7%) lived alone; and 50 (67%) lived in nursing homes. Half were foreign-born and were commonly from Italy, Ireland, or Canada. Mean number of years of education was 11, with a range of 1 to 20 years. Two had doctoral degrees. Of the 74 centenarians for whom we have neuropsychological data as well as medical data, 71 underwent face-to-face neuropsychological evaluation by Dr Silver. Detailed neuropsychological data for the other three were collected from a comprehensive cognitive, functional interview with the family, from medical records, and from other caregiver sources.
Neuropsychological Measures
Test battery.
In choosing the test battery, efforts were made to include well-normed tests. In recent years, the availability of normative data for older age groups has increased; eg, the new version of the Wechsler Adult Intelligence Scale (WAIS) (26) includes norms up to age 89, whereas previously the cutoff was 74. However, norms for nonagenarians are still rare, and norms for 100-year-olds are virtually nonexistent (9, 10).
The MDRS (27) was chosen as the major test in the battery. It assesses five domains of neurocognitive function, ie, attention, initiation/perseveration, construction, conceptualization, and memory. It omits, however, other domains that are important in dementia assessment, such as confrontation naming, drilled auditory learning, and recall after a significant delay, which were assessed by additional tests (Table 1) (27–39).
|
Special modifications.
An important consideration was whether the tests chosen could realistically be administered to centenarians. Fatigue, slow processing, and hearing and vision impairment must be taken into consideration (4, 40) in testing the extremely old. Forty-three percent of the study subjects had vision or hearing impairment that interfered with testing, and combined vision and hearing impairments were experienced by 25%.
To accommodate for visual problems, all test stimuli were enlarged. For the hearing impaired, an amplifier was a standard part of the testing equipment, and family often contributed by conveying information to the subjects because their voices and vocabulary were familiar.
All testing was done in the centenarians’ residences to spare them the strain of being transported to an office. If a subject tired, testing was stopped and completed at another time. Some subjects were able to tolerate 3 hours of testing at one sitting, others only 1/2 hour. In addition to fatigue, psychomotor slowness affected the time required for testing.
English language limitations also affected the testing, and family members often participated as translators. Language limitations also had to be considered in interpretation of scores, eg, conceptualization scores because this cognitive domain is likely to measure at a lower level when English is the second language and/or when educational level is low (41). Lower scores on executive functioning tests, such as verbal fluency and sequencing tests, which are thought to decline with normal aging, were also interpreted cautiously (41).
Determining diagnosis.
Diagnoses for the centenarian subjects took into consideration all the neuropsychological data, functional data, medical data, and history. Specific cutoff scores were not used because of their unavailability for this group and because of the need to look at the entire functional picture, plus education, native language, medical illness, and so on. A CDR score, according to the CERAD (39) was then generated from the review of these data.
Neuropathological Assessment
Two neuropathological methods were used to evaluate Alzheimer changes, consisting of neuritic amyloid plaques and neurofibrillary tangles in the centenarian brains. As part of the neuropathological assessment, each brain was also evaluated for other neurodegenerative diseases or alterations that could contribute to dementia, including macro- or microscopic infarcts or white matter rarefaction.
The Neuropathology Task Force of the CERAD has established a protocol for postmortem examination of brains from demented as well as from nondemented patients (42). Gross examination includes the relative degree of cortical atrophy and ventricular enlargement with note of presence or absence of hippocampal and entorhinal atrophy, pallor of the pigmented brain stem nuclei, the substantia nigra, and the locus caeruleus, the degree of atherosclerosis in the cerebral blood vessels, and the size and distribution of infarcts and hemorrhages.
Method 1: CERAD protocol.
Microscopic examination included a minimum of five anatomic regions from the frontal, temporal, and parietal lobes as well as the hippocampus, entorhinal cortex, amygdala, and substantia nigra. Modified Bielschowsky silver-stained tissue sections were examined for neurofibrillary tangles and senile plaques, which are of two types, neuritic plaques and diffuse plaques. A semiquantitative assessment (sparse, moderate, or numerous) was made of neocortical neuritic plaques from the area of maximum density. This plaque score was correlated with the clinical impression (dementia or no dementia) to determine the degree of certainty of AD (definite, probable, possible, or no evidence of AD). In the centenarian group, a diagnosis of probable or definite AD required evidence of clinical dementia plus moderate (correlating with probable AD) to frequent (correlating with definite AD) numbers of neuritic plaques. Possible AD is suggested by either the combination of dementia and sparse neuritic plaques or moderate to frequent numbers of plaques in the absence of clinical dementia.
Method 2: Braak and Braak staging.
The NIA–Reagan Institute criteria for AD have built on the CERAD approach by including a staging method of neurofibrillary tangles (43, 44). Braak and Braak (17) described a topographic staging method of neurofibrillary tangles defined by six hierarchical stages (I–VI). This staging protocol is a model for the chronology of neurofibrillary tangle accumulation in an individual. A characteristic distribution of neurofibrillary tangles is used to identify six stages in a range of nondemented and demented brains. The earliest two stages (I and II) are characterized by few (I) or numerous (II) accumulations of neurofibrillary tangles in the entorhinal cortex, with possibly rare tangles in other areas of the brain. In the next stages of severity (III and IV), there are greater numbers of neurofibrillary tangles in the entorhinal cortex plus involvement of the hippocampus and possibly a few cortical tangles. In stages V and VI, there is not only severe involvement in the entorhinal cortex and the hippocampus but there are many tangles in the neocortex, with stage VI representing the most pronounced changes, which correspond to fully developed AD.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONDefining Disease-Free Aging in...METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
1). Of the dementia categories, 10 subjects had mild dementia (CDR = 1), 22 had moderate dementia (CDR = 2), 14 had severe dementia (CDR = 3), 4 had profound dementia (CDR = 4), and 6 had terminal dementia (CDR = 5) (Table 2).
|
Neuropathological studies of these 14 brains have been performed according to a standardized protocol (Table 3). These results demonstrate that some subjects can live to extreme old age without significant accumulation of pathological markers of AD, as has been reported in other studies (5, 7). It is also evident that not all dementia in this age group is AD and that some dementias have other clear pathological etiologies (eg, Pick’s disease), although the etiology of still other dementia symptoms is less clear. Neuropsychological data for four centenarians with no dementia correlated well with neuropathological findings of minimal AD changes. For six centenarians with neuropsychological finding of dementia, neuropathological results correlated positively, ie, four centenarian brains met CERAD criteria for definite AD, one centenarian had Pick’s disease, and one had hippocampal sclerosis that accounted for dementia. Thus, in 8 of 14 cases, neuropsychological results correlated with neuropathological findings.
|
Although the combination of hippocampal and AD changes in K.J.’s brain may explain her dementia symptoms, how do we explain absence of dementia in F.F., who also had left hippocampal sclerosis but no dementia by neuropsychological evaluation? Could a functional reserve, a brain reserve that delays the functional expression of pathologic changes, described by Mortimer (23) and others (28), have allowed F.F. to compensate for left hippocampal damage and other brain changes? F.F., a Phi Beta Kappa with a college degree in engineering, was a life-long learner. He continued to read avidly despite severe vision impairment and to give talks on scientific subjects until shortly before his death. Furthermore, he was an avid violinist who played in groups most of his life, and this complex activity continued to challenge his brain (K.J. attended school in Sweden only until age 14 and was neither an avid reader nor a musician). It is possible that F.F.’s continued learning and musical activities helped facilitate compensatory functional brain reserve.
Clinical symptoms of dementia were present in two additional subjects (M.V., L.B.). M.V., with moderate dementia, had sparse neuritic plaques and a Braak and Braak stage of II. Vascular changes included severe cerebral arteriosclerosis with a solitary remote infarct in the occipital lobe that alone would not be expected to account for her clinical deficits. L.B., with mild dementia, had sparse neuritic plaques and a Braak and Braak stage of IV. Vascular lesions were minimal. L.B.’s brain contained three microscopic infarcts in the parietal cortex, hippocampal formation, and thalamus and a small infarct (<1 cm in greatest size) in the cerebellum. Neither vascular nor AD changes alone could account for dementia symptoms. However, both M.V. and L.B. had a sudden onset of symptoms, not typical of AD, and both scored positive for depression on the GDS. Depression was thought to be a significant etiological factor. Other contributing factors may have been medical illness and sensory impairment, which is known to correlate highly with dementia in the oldest old (47, 48).
L.B., a PhD in psychology, would be expected to have acquired significant cognitive reserve. According to history, at age 108, she did not have significant cognitive deficits. Her rapid decline began with an episode of shingles. When evaluated at age 109, the aggregate of many predisposing factors (medical illness, severe hearing deficit, and depression) may have contributed to clinically evident dementia. Furthermore, she died a few months after testing. We were not examining her in the disease-free state that lasted for 8 years beyond her 100th birthday but in a period of acute morbidity just before death.
The six subjects who scored CDR = 0 on neuropsychological evaluation (J.D., A.M., F.F., L.B.L., A.B., J.L.) had sparse to frequent neuritic plaques. Neurofibrillary tangle staging revealed Braak and Braak stages of I to IV. However, for two of these subjects, J.L. and L.B.L., neuropathological findings were more extensive than neuropsychological studies would suggest. Although J.L. had sparse to moderate neuritic plaques, neurofibrillary tangles were rated as Braak and Braak stage III to IV (Gold and colleagues (16) have suggested that, in centenarians, stage IV represents fully developed AD.) L.B.L. had sparse to frequent neuritic plaques, raising the possibility of AD on neuropathological grounds. Additional findings included numerous remote cerebral infarcts involving the superior frontal, primary, and somatosensory cortices and superior and inferior parietal lobes, all in the right hemisphere. There was severe cerebrovascular arteriolosclerosis with lacunar infarcts in the globus pallidus and putamen.
These discrepancies raise the question of why significant pathological markers were not clinically expressed as dementia. As in the case of F.F., a functional reserve may have played a role. J.L. did not have much formal education, but she elected to have cataract surgery in her 90s because reading was so important to her, and at 103, she still read constantly. She continued to make major decisions in the family and balance her checkbook without a calculator. Another possible protective factor was her living situation; she had resided in the same apartment for many years, living independently, with a grandson in the same apartment and a daughter a block away.
Similarly, L.B.L. continued to engage in challenging activities, teaching her paralyzed hand to function during the last few months of her life. She had taken up painting in oils and watercolors in her 90s and, like J.L., was an avid reader. At age 100, she wrote a book about her experiences in the Holocaust and gave a speech on the subject in front of a large group.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONDefining Disease-Free Aging in...METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
The discrepancies between daily functioning and neuropsychological and/or neuropathological results in the cases of L.B.L. and J.L. pose a dilemma in trying to define disease-free aging. Is disease-free aging the ability to maintain a reasonable level of daily functioning? Or does it mean avoidance of testable cognitive impairment? Or do we define disease-free aging as the absence of pathological markers of AD or other brain diseases? As reported earlier, there are centenarians who meet the gold standard, ie, they have no dementia on testing, no functional deficits, and an absence of pathological markers, but how do we interpret discrepancies between neuropsychological and neuropathological findings? This is further complicated by the fact that different criteria may have been used by different researchers to arrive at their conclusions.
These studies of centenarian neuropsychological and neuropathological correlations highlight the importance of the study of centenarian brains in the understanding of cognitive aging. Contrary to a common belief that, by the age of 100, everyone would be demented, we found that 6 of 14 centenarians were not clinically demented. Nine of the centenarian brains did not meet criteria for definite of even probable AD by CERAD criteria. This underscores the mild nature of neuritic changes among this group, which included both demented and nondemented subjects. We discovered that a 100-year-old brain can be essentially free of the plaques and tangles associated with AD and that some individuals can live to 100 without significant cognitive decline. Although dementia is associated with aging, this and other studies (5, 7) suggest that dementia is not inevitable with aging, at least into the 11th decade.
Another notable finding is that not all dementia in centenarians is AD. Depression, sensory impairment, medical illness, and environmental factors can all affect cognitive functioning to a significant degree, as can be seen in subjects such as L.B. In this case, more intact recognition memory than recall and absence of word finding difficulties distinguished between the dementia of depression and organically caused dementia, a similar pattern to that in younger adults. However, normal aging changes, such as slower cognitive processing, set shifting difficulties, and mild short-term memory decline may have been exacerbated by depression, which may explain why dementia related to depression occurs more frequently in older than in younger adults (54, 55).
The finding that the presence of markers of neurodegenerative disease, including significant AD changes, does not always predict clinical symptoms of dementia raises questions about the possibility of a compensatory functional reserve that delays clinical expression of brain changes.
IMPLICATIONS
The definition of disease-free aging in extreme old age remains unclear. Is disease free the lack of clinical dementia symptoms or the absence of pathology at the cellular level? As for defining usual cognitive aging, there are those who argue that centenarians are a select population, even a supernormal population, who cannot be exemplars of what is usual in the aging process for the general population. It may even be difficult to find norms or commonalities within the 100-year-old cohort because of their diversity and variety of life experience. However, with the population of centenarians growing at such a rapid rate, a much larger population group may bring researchers closer to discovering what is usual for this group.
The discrepancies between neuropsychological and neuropathological findings in some centenarians suggest that there are possible risk and/or protective factors for AD and other contributors to dementia, eg, hippocampal sclerosis. Other studies have suggested that brain infarction may lead to clinical expression of AD symptoms despite a relatively lower burden of Alzheimer changes (56). Two centenarian subjects (M.V., K.J.) had minimal infarcts that in themselves would not account for dementia symptoms but may have contributed to expression of dementia with few AD changes. (However, a third subject with vascular changes, L.B.L., had no dementia according to neuropsychological testing.) M.V.’s and K.J.’s significant hearing impairments may also have played a role in expression of dementia. It is known that hearing impairment is a major risk factor for development of cognitive deficits (47), which suggests the importance of the development of better hearing technology to decrease dementia risk.
By examining the spectrum of clinical presentations and their neuropathological correlates in the extremely old, we have demonstrated the importance of diagnosing reversible vs. nonreversible cause of cognitive impairment so that reversible conditions, such as depression and certain medical illnesses, can be treated. Although depression increases risk of dementia symptoms in younger age groups (54, 55), older adults are more at risk. However, depression is less likely to be detected in this more vulnerable group, highlighting the need for greater education and better screening programs so that reversible dementia of depression can be treated.
The finding that pathological brain changes are not always expressed as clinical dementia has suggested that factors that protect against dementia should be studied further. The creation of optimal environments would benefit not only those with sensory impairment but also those with mild cognitive deficits, such as J.L., who can continue to function in a familiar, supportive, and user-friendly environment.
A potent protective factor against cognitive decline and dementia may be the development of a functional brain reserve, which may have been a factor in preserving cognition in the context of AD changes and hippocampal sclerosis in centenarian subjects. Additional study of the concept of functional reserve is called for because of its potential role in developing preventive interventions and programs that may delay the expression of dementia symptoms. Centenarians, like F.F., who continued to study political theory into his second century of life, may give us clues as to what kinds of activities or interventions facilitate the strengthening of this potentially preventive factor.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONDefining Disease-Free Aging in...METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Received for publication August 1, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONDefining Disease-Free Aging in...METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  V. Haroutunian, M. Schnaider-Beeri, J. Schmeidler, M. Wysocki, D. P. Purohit, D. P. Perl, L. S. Libow, G. T. Lesser, M. Maroukian, and H. T. Grossman Role of the Neuropathology of Alzheimer Disease in Dementia in the Oldest-Old Arch Neurol, September 1, 2008; 65(9): 1211 - 1217. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. T Perls The different paths to 100 Am. J. Clinical Nutrition, February 1, 2006; 83(2): 484S - 487S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Peters Ageing and the brain Postgrad. Med. J., February 1, 2006; 82(964): 84 - 88. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. G. Ory and M. Chesney Aging and the Life-Course: Advancing Psychosomatic Medicine Research Psychosom Med, May 1, 2002; 64(3): 367 - 369. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
