THE AGEING BRAIN

THE AGEING BRAIN

THE NEUROBIOLOGY AND
NEUROPSYCHIATRY OF AGEING

Edited by

PERMINDER S. SACHDEV
This edition published in the Taylor & Francis e-Library, 2005.
“To purchase your own copy of this or any of Taylor & Francis or Routledge’s
collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.”
Copyright © 2003 Swets & Zeitlinger B.V., Lisse, The Netherlands

All rights reserved. No part of this publication or the information contained herein
may be reproduced, stored in a retrieval system, or transmitted in any form or by
any means, electronic, mechanical, by photocopying, recording or otherwise, without
written prior permission from the publishers.

Although all care is taken to ensure the integrity and quality of this publication and
the information herein, no responsibility is assumed by the publishers nor the author
for any damage to property or persons as a result of operation or use of this publica-
tion and/or the information contained herein.

Published by: Swets & Zeitlinger Publishers

www.szp.swets.nl
ISBN 0-203-97097-7 Master e-book ISBN

ISBN 90 265 1943 5 (Print Edition)

Contents

ACKNOWLEDGEMENTS ix

Section I Introduction 1

CHAPTER 1: THE AGEING BRAIN 3

Perminder S Sachdev

CHAPTER 2 POPULATION AGEING, HUMAN LIFESPAN AND

NEURODEGENERATIVE DISORDERS: A FIFTH EPIDEMIOLOGIC
TRANSITION 11
G Anthony Broe

Section II Characteristics of the ageing brain 33

CHAPTER 3 STRUCTURAL CHANGES IN THE AGEING HUMAN BRAIN 35

Jillian J Kril

CHAPTER 4 STRUCTURAL NEUROIMAGING OF THE AGEING BRAIN 49

Jeffrey CL Looi and Perminder S Sachdev

CHAPTER 5 NEUROPHYSIOLOGICAL, SENSORY AND MOTOR CHANGES

WITH AGEING 63
Stephen R Lord and Rebecca St George

CHAPTER 6 COGNITIVE CHANGES AND THE AGEING BRAIN 75

Helen Christensen and Rajeev Kumar

CHAPTER 7 AGEING OF THE HUMAN BRAIN AS STUDIED BY

FUNCTIONAL NEUROIMAGING 97
Julian N Trollor and Perminder S Sachdev
vi CONTENTS

CHAPTER 8 NEUROENDOCRINE ASPECTS OF BRAIN AGEING 139

George A Smythe

CHAPTER 9 CEREBROVASCULAR SYSTEM AND THE AGEING BRAIN 153

Valendai K Srikanth and Geoffrey A Donnan

Section III Factors influencing brain ageing 171

CHAPTER 10 THE MOLECULAR BASIS OF ALZHEIMER’S DISEASE

AND FRONTOTEMPORAL DEMENTIA 173
John BJ Kwok and Peter R Schofield

CHAPTER 11 OXIDATIVE AND FREE RADICAL MECHANISMS IN BRAIN

AGEING 187
Judy de Haan, Rocco C Iannello, Peter J Crack, Paul Hertzog and
Ismail Kola

CHAPTER 12 THE ROLE OF NUTRITIONAL FACTORS IN COGNITIVE

AGEING 205
Janet Bryan

CHAPTER 13 THE BRAIN RESERVE HYPOTHESIS 223

Peter W Schofield

Section IV Clinical interface 241

CHAPTER 14 WILL WE ALL DEMENT IF WE LIVE LONG ENOUGH? 243

Carol Brayne

CHAPTER 15 DETECTING ALZHEIMER’S DISEASE AT THE

PRE-SYMPTOMATIC STAGE 259
Gary W Small

CHAPTER 16 PARKINSONISM AND AGEING 275

John GL Morris, Mariese A Hely and Glenda M Halliday

CHAPTER 17 AGE VARIATION IN THE PREVALENCE OF DEPRESSION:

ARE STUDY FINDINGS MEANINGFUL? 283
John Snowdon

CHAPTER 18 VASCULAR DEMENTIA 299

Perminder S Sachdev
CONTENTS vii

CHAPTER 19 CONCLUSION 323

Perminder S Sachdev

CONTRIBUTORS ADDRESS LIST 327

SUBJECT INDEX 333

AUTHOR INDEX
Acknowledgements

The seed for this book was sown with the formation of The Ageing Brain
Program at the University of New South Wales in 1998, and the early sprout
appeared in 2000 at the International Conference on the Ageing Brain held at
the Scientia, University of New South Wales, Sydney. The book is, of course,
more than the Conference, and its diverse foliage is the dedicated work of
many scientists and scholars. I am extremely grateful to all the authors for
that extra effort that made each of the chapters a significant contribution.

The editing of a book is a labour of love that demands doggedness and com-
pulsive persistence. The latter qualities were brought to this work with meas-
ured good humour by Angela Russell, who undertook the tasks of editing and
compiling. If she sometimes annoyed the contributors with her deadlines and
diligent proofreading, the final manuscript will more than compensate for it.
She was assisted in this task by the quiet and behind-the-scenes contribution
of Wanda Schinke. The flair of Joanna Christie was an important determinant
of the success of the Conference.

The planning of this book, and its intellectual content, were influenced by
many colleagues to whom I am extremely grateful. I would like to make
particular mention of Sam Aroni, Henry Brodaty, Tony Broe, Felicia Hup-
pert, Jeffrey Looi, Gary Small, Julian Trollor, Michael Valenzuela and Xing
Li Wang who were generous with their suggestions and time. I found, in the
publishers of this book, a rather indulgent group of professionals, led by
Arnout Jacobs, who let many deadlines go past with little more than gentle
reminders.

For the undisturbed small hours of the morning that the writing took me
into, I am in debt of my family — my beautiful wife Jagdeep and our lovely
daughters Nupur and Sonal. They have provided the environment which has
continued to nurture me through all my academic travails.

My research into neuropsychiatric disorders of the elderly has been supported

generously by the University of New South Wales and the National Health
x ACKNOWLEDGEMENTS

and Medical Research Council of Australia. Additional support has been pro-
vided by the Rebecca Cooper Foundation, the Brain Foundation, the Fairfax
Foundation and Pfizer Inc. None of these organizations has any vested inter-
est in the intellectual content of the book or any commercial interest in it.

Perminder S Sachdev
 SECTION I

INTRODUCTION
Chapter 1

THE AGEING BRAIN

Perminder S. Sachdev

According to Hesiod, a Greek philosopher, the history of mankind could be

divided into five epochs. The first was the Age of Gold in which mortals never
aged and peace and happiness were pervasive. This was followed by the Age of
Silver in which childhood lasted a hundred years but adulthood was transient.
The third age, in which Hesiod lived, was the Bronze Age, which was a time of
greed, corruption, injustice and violence. When this ended, Zeus created the
Heroic Age in which the world was populated by demigods. Then came the Iron
Age, in which we now live. Hesiod wisely predicted that this would be the age
of violence, the love of profit, and an increasingly decadent lifestyle. Hesiod
predicted that Zeus would be particularly incensed by the lack of honour shown
to the elderly by the young, and by children not repaying their parents for the
nurturance they received. Zeus would then create a new and idyllic Age.
A hallmark of our Age is also the belief that future Ages are of our own
making. Few would disagree that the salient characteristic of a golden age of
the future would be eternal youthfulness, or at least youthfulness until the
time of delayed but sudden death. This may explain our preoccupation with
ageing. The images of ageing we confront on a daily basis are contrasting in
nature. For those of us who are in mid-life, ageing represents a relentless ero-
sion of our vitality. There are the obvious reminders in the greying hair, the
balding scalp, the slight stiffness in the joints, and the small lapses of memory.
The death notices in the newspaper become noticeable. A sudden dread fills
our hearts as we witness our parents succumb to the travails of senescence.
Yet we still hope to age like fine wine, accumulating Talmudic wisdom with
our years. Individuals, like the late Madame Jean Calment of France, remind
us that we could be living independently well into the 12th decade of our
lives.1 We wonder if the tools of modern biology will uncover the mysteries
of ageing and help control, if not reverse, it. We look with wonderment at the
genome project and the vast worldwide army of biomedical scientists. Our
dread is inter-mixed with awe and expectation.
4 THE AGEING BRAIN

More than any other organ of the body, we are concerned with the ageing
of the brain. The ageing brain must be considered a special case within the
domain of ageing. While age-related changes in the brain in general parallel
those of the body, there are important exceptions. Brain diseases that are
usually regarded as concomitants of old age, such as Alzheimer’s disease
(AD) and Parkinson’s disease (PD) are sometimes seen in the young, and most
elderly individuals manage to evade them. It is not uncommon to see a very
active mind in a frail old body. Our examination of the ageing of the brain
must therefore occur in the context of a neurobiological understanding. The
ageing brain is a special case also for social reasons. An epidemic of dementia
is upon us, and governments in the developed world are preoccupied with the
impact this will have on the society of the future. As Professor Broe states
later in this book, we are in the Age of Neurodegenerative Disorders, and
insights into the mechanisms of ageing are urgently needed. .
In the face of this dread, we must take heart in the pace of neuroscientific
research, which has been breathless indeed. Let us consider a few examples.
Most studies both in vivo and post mortem, suggest shrinkage of the adult
brain as it ages, with a reported reduction of about 5% in brain weight per
decade after the age of 40 years.2 This change is not uniform, however, with
the prefrontal regions being affected more than the temporal and parietal
neocortex. In the subcortical regions, the neostriatum atrophies moderately
with age, while the globus pallidum and thalamus are relatively spared.3
What we do not understand are the reasons for this variation. Why is the
substantia nigra, for example, more susceptible to age-related degeneration
than the thalamus? What are the determinants of hippocampal degeneration
seen in ageing brains? Questions such as this may be the keys to understand-
ing the physiological processes involved in brain ageing. It is quite likely that
the mechanisms of neural ageing are the same as for the rest of the body.
There must also be important differences. A large part of the human genome
is involved in brain development, suggesting that a great complexity must be
fathomed.
It was thought for many years that the changes in brain volume seen in
ageing were a consequence of age-related neuronal loss.4 This notion was
so well accepted that it had entered lay parlance. Recent studies using bet-
ter stereological methods have shown that this may not be true, and in fact
most brain regions do not suffer an age-related neuronal loss.5 If there is a
sparing of the total number of cortical neurones, what is the basis of loss of
cortical volume with ageing? The hippocampus has been studied extensively
to understand this, and it has been shown that its functional organization is
altered with ageing. This is related to alterations in connectivity, because of
reductions in dendrites and synapses. In both rodents and humans, changes
have been reported in dendritic arbor, spines and synapse morphology that
could impact on the function of hippocampal circuits but would not be
reflected as neuronal loss.6 This is functionally important as the most cog-
nitively impaired aged rats demonstrate the greatest degree of abnormality.
THE AGEING BRAIN 5

The number of synapses can be judged by the density of receptors in the

molecular layer as has been shown for the glutamate NMDA receptor which
plays a critical role in mechanisms of plasticity comprising the cellular basis
of learning and memory.7
The physiological implications of this change with ageing have been stud-
ied from many perspectives, one of which is long-term potentiation (LTP).
This is a functional change in synaptic transmission secondary to neuronal
stimulation, and has a role in memory functioning. The stimulation necessary
to induce peak LTP, and the maximal potentiated response attained, are the
same in young and old brains, but LTP decays to the prepotentiated baseline
levels more rapidly in aged subjects — a possible reason for the “forget-
ting” experienced by the elderly. Many aspects of synaptic transmission are
unaffected by age and there may even be compensatory changes. Functional
imaging studies show that aged brains are less efficient in the processing of
information, tending to recruit more extensive networks of neurones.8 This
may, however, be a correctable change, since the brain is known to retain
much of its plasticity despite age, and presents a potential for intervention.
The cognitive changes associated with ageing are the subject of intense
research. It is reassuring that crystallised intelligence remains intact with
age, although some cognitive abilities do show a gradual reduction. Age-
ing causes a decline in information-processing resources, such as working
memory capacity, attentional regulation and processing speed. Ageing results
in greater intra-individual and inter-individual variation in performance. The
change in these resources must be understood from a neurobiological perspec-
tive. It is interesting that the onset of the decline is relatively early in life, and
to some extent parallels the decline in physical and sensory functions. The
relationship with structural brain change is far from perfect, and can only
be demonstrated in those with a pathological degree of impairment. It may
be also be interesting to examine the cognitive decline using computational
theories of neuronal function.9 Does the change in resources lead to deficient
neuromodulation and increased neural noise, i.e. haphazard activation during
neuronal information processing? There is evidence that mental representa-
tions in the elderly are less distinctive. Events that happen in the course of a
day are less distinctly remembered by older individuals, suggesting that they
may be processing information less elaborately. Neuro-imaging studies show
that the elderly are more likely to activate both hemispheres for tasks that are
lateralised in young individuals.10 One proposed mechanism for the increased
neuronal noise with age is reduced neuronal responsivity due to a declining
dopaminergic modulation.9 There are obviously other possibilities, which
prompt for a multi-level approach to the problem of cognitive ageing.
When one examines the risk factors for cognitive decline with age, one is
confronted with the question: how much of the change is because of pathol-
ogy in the brain? This is an issue not easy to settle. An ageing brain accu-
mulates pathology that may be due to cerebrovascular disease or systemic
diseases with their secondary brain effects. This may account for some of
6 THE AGEING BRAIN

the age-related deficits that are likely to be misattributed to ageing-related

changes. As an example, brains of elderly individuals frequently show hyper-
intense signals on T2-weighted magnetic resonance imaging (MRI), which has
been the subject of hundreds of studies.11 Are these findings always indica-
tive of pathology, or can they represent normative ageing-related changes?
What role do these “lesions” play in the development of cognitive change
and psychiatric disorders in late life? It is necessary to pose such questions to
understand the nature of the “normal” ageing process.
We have seen impressive advances in our understanding of neurodegenera-
tive diseases in the last two decades, and are now at the threshold of effective
treatments. Much work however remains to be done. Epidemiological studies
have revealed many risk factors that have to be explained in terms of patho-
physiological processes, and major gaps in this understanding remain. Genetic
factors still explain only a minority of cases of Alzheimer’s disease (AD). The
great divide between neurodegeneration and vascular pathology is no longer
an unbridgeable gulf, but the physiological basis for this association is yet
to be understood. The argument whether AD is an extreme form of ageing is
unresolved, and some do believe, as Carol Brayne argues in this book, that
if we live long enough, all of us will succumb to AD. We know a great deal
about other causes of brain degeneration, but the reasons why one brain with
fronto-temporal degeneration produces Pick bodies and not another remain
elusive. Why is it that a particular region of the neocortex is preferentially
affected in some dementias such as semantic dementia, progressive aphasia,
etc.? This is a field in which the contributions of clinicians, epidemiologists
and neuroscientists have gone hand in hand, and the future lies in the contin-
ued cooperation between disciplines.
If we are to influence the ageing process, it is necessary that we understand
the underlying molecular mechanisms. The frontier of the biochemistry of
ageing, although yet to be conquered, is witness to many raging battles. As a
consequence, terms like free radicals, heat shock proteins and nerve growth
factors have become household words. Some of these theories have inextri-
cably linked the brain with the rest of the body. As an example, the role of
corticosteroids in the stress response, and its influence on brain structures
has provided a link between psychology and biology.12 Brain regions that
are important for learning and memory processes are particularly sensitive
to stress hormones. The hippocampus has a high concentration of adrenal
steroid receptors. Stress can thereby impair memory acutely; and chronic or
repeated stress can lead to atrophy of dendrites and reduced neuronal connec-
tions. If prolonged this change can become irreversible and loss of neurones
results. It has also been shown that early stress, such as prolonged separation
of rat pups from their mothers, may lead to a chronic over-reactivity to stress
in these animals. This may result in accelerated brain ageing. Other hormones
such as oestrogen, growth hormone, melatonin, testosterone and dehydroe-
piandrosterone are being examined for their role in reversing some aspects of
ageing. Growth or trophic factors abet these.
THE AGEING BRAIN 7

The brain is intricately linked with the immune system, and age-related
changes to the immune system have been of special interest to neurobiologists.
The function of T-cells and their ability to proliferate declines with age. The
T-cells produce powerful chemicals known as lymphokines, which mobilise
mediators of the immune response. The effects of age on these lymphokines
are variable, with rise in some and fall in others. It is not known how this
may be linked to neuronal function.
In this age of genomics, some of the causes of brain ageing are being sought
in genetic factors. Most of the progress in neurogenetics has been in discov-
ering genes for various neurological diseases, including those that affect the
elderly. There have been exciting developments in AD, with the discovery of
three genes that cause early-onset AD. However, this accounts for the disor-
der in but a small proportion of AD patients. The discovery of the tau gene in
fronto-temporal dementia has raised the question of the relationships between
the different genes, and what pathways may be shared in neurodegenerative
disorders. The pace of this research is likely to increase as animal models are
established.
Recent research has shown that the expression levels of many genes related
to neuronal signalling, plasticity and structure are altered with ageing.13 For
example, the expression of certain proteases, such as prolyl oligopeptidase
and caspase-6, is up-regulated in the aged brain. These proteases play essen-
tial roles in regulating neuropeptide metabolism, amyloid precursor protein
processing, and neuronal apoptosis, and are likely contributors to brain age-
ing. It is interesting that some of these changes in gene expression can be
reversed by environmental enrichment,14 providing hope for intervention.
The mapping of the human genome, and the recognition that a large number
of genes are involved in brain development, has opened up exciting opportu-
nities for understanding the molecular basis of brain ageing.
It would be important to find out if there are a few major genes that deter-
mine ageing, or is it the result of the cumulative effect of changes in many
genes? Is ageing the result of defects in DNA repair that gradually accumulate?
Are there some genetic modifications that can delay, if not stop or reverse
the processes of brain ageing? Does dietary restriction delay ageing through
genetic factors?15 Attempts have recently been made to apply gene transfer
technology to protect neurones from death following neurological insults.16
It is conceivable that gene therapy in the future may be able to protect the
nervous system from ageing. Transgenic intervention could be in order to
over-express a particular gene to protect against decline of its product in old
age, or gene therapy could target a discretely damaging event highly likely to
occur in the elderly. Technologies for the delivery of genes into neurones to
maintain function and protect against injury are being developed.
Genomics is likely to be complemented by the newly developed science
of Proteomics,17 as there are many more proteins in the human body than
can be accounted for by the number of genes recognised on the genome. The
nearly 30,000 genes have a complement of nearly 300,000 proteins, and each
8 THE AGEING BRAIN

of the 200 cell types in the body has a different set of proteins. The proteins
produced by the cells at any particular time are moderated according to
the biological needs of the body, and are influenced by the disease process
present. Proteomics therefore permits the identification of proteins associated
with particular diseases. This will assist diagnosis, as is already the case in
AD, and speed the development of new treatments. It also offers the exciting
possibility that drugs may be tailored the to individual patient, opening up
an era of personalised medicine.
The treatments of neuropsychiatric disorders of the elderly are likely to
look very different in the future, as suggested by the above developments. In
many respects, the future is already here. Depressive disorders have recently
seen the introduction of two novel treatments: transcranial magnetic stimu-
lation and vagus nerve stimulation. Parkinson’s disease patients worldwide
are benefiting from deep brain stimulation. Targeted drugs are increasingly
being developed for specific receptors. Stem cells promise to open up a new
era in therapy. In fact, recent findings suggest that a decline in the numbers
and plasticity of stem cells may contribute to ageing itself.18 It is likely that
methods will be developed to tweak the stem cells already in the brain, or
introduce new ones, to replace lost or dysfunctional cell populations.
The above developments reveal the rapidity with which new information is
being acquired and old orthodoxies challenged. However, a glorious ageless
society is not upon us yet. In the medium-term, our goals as a society must
be limited. We can start by emphasising the positive aspects of old age. Some
people may find this a difficult concept to grasp, and yet for thousands of
years, societies have valued age and even venerated it. The wisdom of old age
is difficult to quantify, but recent research showed that on a rational choice
task, 70 year old subjects performed much more consistently than those 50
years younger.19 With the inevitable ageing of our populations, we have lit-
tle choice but to make old age productive, healthy and enjoyable. Much of
this will be achieved through social and political change and not medical
advances. An increasing number of healthy older people can make a signifi-
cant contribution to the lives of younger generations. Age can help temper
and direct the energy of the young.
Medical science does not, in the near future, hope to conquer ageing.
It can have a more modest goal, however, in delaying the onset of late-life
dysfunction. Old age is characterised by an array of ageing-related diseases,
which include cardiovascular disease, dementia, sensory deficits, Parkinson’s
disease, diabetes, osteoporosis and incontinence. The mere delaying of the
onset of some of these will have a major public health impact. For instance,
a delaying of the onset of Alzheimer’s disease by five years will halve the
prevalence of the disorder. It is this promise that is prompting a burgeoning
industry of health promotion. Low-fat labels, cholesterol free diets, folic acid
supplementation, aspirin prophylaxis, anti-oxidants and organic foods are
more than passing fads. In this rush toward a healthy old brain, it is difficult
to separate established scientific facts from overvalued ideas. A few messages
THE AGEING BRAIN 9

do seem to have sufficient empirical basis. We can protect the brain some-
what if we control hypertension early and effectively, and attend to other
cerebrovascular risk factors such as diabetes, smoking, high cholesterol and
obesity. We should aim for moderation in our use of alcohol, and perhaps
try to restrict it to red wine, while we refrain from using illicit substances.
Whether we will benefit from using anti-oxidants or anti-inflammatory drugs
remains to be established. The use of folic acid supplementation to reduce
serum homocysteine levels is again not established as an epidemiological
health measure20 but is increasingly popular. Also without sufficient scientific
backing, the use of a daily multivitamin tablet that does not exceed the RDA
of its components makes sense for most adults, given the greater likelihood of
benefit than harm and the low cost.21 The use of vitamin E at 400 IU per day
in middle and old age by those at risk of vascular disease can also be recom-
mended.21 Stress, no doubt, is bad for the body and the brain, and has been
linked with psychiatric and cognitive disorders, and we should unequivo-
cally recommend stress-reduction strategies to our patients. The action of
stress on neurones is most probably through the glucocorticoid cascade. This
response can be modified by environmental manipulation as early as in the
neonatal period,22 and continuing on into later life. It is interesting that this
manipulation of the adrenocortical axis can safely and effectively be brought
about by a psychologist.22 The promotion of other hormones, such as growth
hormone, melatonin, DHEA, pregnenolone, testosterone, oestrogen and pro-
gesterone, as elixirs of youth is without unambiguous scientific basis.
A study from Boston23 showed that exercise can strengthen muscles,
improve mobility, and reduce frailty even among 90-year-old individuals.
The same may be true for the brain, which harbours a significant potential for
plastic change well into old age.24 Another analogy to be drawn with muscles
is that the brain has a reserve than can be influenced by mental activity, and
serves to protect the individual from age-related changes.25
A number of studies have reported that higher educational and occupa-
tional levels, mental activity and high intellectual performance are protective
factors for dementia. These findings, and those relating to nutritional factors
and stress, promote an agenda for the future that is hopeful, and suggest
interventions at the population level that should begin now without awaiting
breakthroughs in the understanding of molecular processes. It is important
to take this message to decision-makers if we are to influence the future of an
ageing society.

References

1. Ritchie K. Mental status examination of an exceptional case of longevity — J.C.

aged 118 years. Br J Psychiatry. 1995; 166:229–235.
2. Kemper T. Neuroanatomical and neuropathological changes during aging and
in dementia. In: Albert M, Knoepfel J, editors. Clinical neurology of aging. New
York: Oxford University Press, 1994; 3–67.
10 THE AGEING BRAIN

3. Trollor J, Valenzuela M. Brain ageing in the new millenium. ANZ J Psychiatry.

2001; 35:788–805.
4. Brody H. Structural changes in the aging nervous system. Interdiscip Topics
Gerontol. 1970; 7:9–21.
5. Wickelgren I. Is hippocampal cell death a myth? Science. 1996; 271:1229–
1230.
6. Hamrick J, Sullivan P, Scheff S. Estimation of possible age-related changes in
synaptic density in the hippocampal CA1 stratum radiatum. Soc Neurosci Abstr.
1998; 24:783.
7. Gazzaley AH, Siegel SJ, Kordower JH, Mufson EJ, Morrison JH. Circuit-specific
alterations of N-methyl-D-aspartate receptor subunit 1 in the dentate gyrus of
aged monkeys. Proc Natl Acad Sci USA. 1996; 93:3121–3125.
8. Almkvist O. Functional brain imaging as a looking glass into the degraded brain:
reviewing evidence from Alzheimer disease in relation to normal aging. Acta
Psychol. 2000; 105: 255–277.
9. Li S-C, Lindenberger U, Sikstrom S. Aging cognition: from neuromodulation to
representation. Trends Cog Sci. 2001; 5:479–486.
10. Cabeza R, McIntosh AR, Tulving E, Nyberg L, Grady CL. Age-related differences
in effective neural connectivity during encoding and recall. NeuroReport. 1997;
8:3479–3483.
11. Pantoni L, Garcia J. Pathogenesis of leukoaraiosis: A review. Stroke. 1997; 28:
652–659.
12. Sapolsky R. Stress, the aging brain, and mechanisms of neuronal death. Boston:
MIT Press, 1992.
13. Jiang CH, Tsien JZ, Schultz PG, Hu YH. The effects of aging on gene expres-
sion in the hypothalamus and cortex of mice. Proc Nat Acad Sci USA. 2001; 98:
1930–1934.
14. Rampon C, Jiang CH, Dong H, Tang YP, Lockart DJ, Schultz PG, Tsien JZ, Hu
YH. Effects of environmental enrichment on gene expression in the brain. Pro
Nat Acad Sci USA. 2000; 97:12880–12884.
15. Weindruch R, Walford RL. The retardation of aging and disease by dietary
restriction. Springfield, IL: Charles C Thomas, 1988.
16. Ogle WO, Sapolsky RM. Gene therapy and the aging nervous system. Mech Age-
ing Dev. 2001; 122:1555–1563.
17. Banks RE, Dunn MJ, Hochstrasser DF, Sanchez JC, Blackstock W, Pappin DJ,
Selby PJ. Proteomics: new perspectives, new biomedical opportunities. Lancet.
2000; 356:1749–1756.
18. Rao MS, Mattson MP. Stem cells and aging: expanding the possibilities. Mech
Ageing Dev. 2001; 122:713–734.
19. Tentori K, Osherson D, Hasher L, May C. Wisdom and aging: irrational prefer-
ence in college students but not older adults. Cognition. 2001; 81:B87–96.
20. Diaz-Arrastia R. Homocysteine and neurologic disease. Arch Neurol. 2000; 57:
1422–1427.
21. Willett WC, Stampfer MJ. Clinical Practice. What vitamins should I be taking,
doctor? New Eng J Med. 2001; 345:1819–1824.
22. Seligman M. Learned optimism. New York: Alfred Knopf, 1991.
23. Levine S. Plasma-free corticosteroid response to electric shock in rats stimulated
in infancy. Science. 1962; 135:795–798.
24. Anstey K. How important is mental activity in old age? Austr Psychol. 1999; 34:
128–131.
25. Schofield P. Alzheimer’s disease and brain reserve. Australas J Ageing. 1999; 18:
10–14.
Chapter 2

POPULATION AGEING,
HUMAN LIFESPAN AND
NEURODEGENERATIVE
DISORDERS: A FIFTH
EPIDEMIOLOGIC
TRANSITION
G Anthony Broe

Introduction

Rapid population ageing, or a rising percentage of older people in the popu-

lation, was a 20th century phenomenon in developed countries and is now
affecting most of the world’s populations, the poor as well as the rich. Coun-
tries as disparate as Australia, Iran, Thailand and Tunisia are approaching or
achieving below replacement fertility levels.1
The immediate cause of world population ageing was fertility decline; this
followed the reduction in infant deaths due to infectious diseases from the
19th century onwards and their substitution by deaths due to adult onset
degenerative diseases in the first half of the 20th century. The infectious
diarrhoeas, influenza and tuberculosis, in children and young people, were
gradually replaced by cardiovascular and lung diseases at older ages. Omran,2
in 1971, referred to this shift in disease patterns as the “Epidemiologic Tran-
sition”; and he described three disease transitions occurring in developed
countries up to the mid-20th century.
This chapter examines population ageing and life span in relation to fur-
ther disease transitions and changing causes of mortality and morbidity later
12 THE AGEING BRAIN

in the 20th century. A progressive delay in age of onset, and a decline in

mortality from the systemic degenerative diseases (such as cardiovascular and
lung diseases) was described as the fourth transition of “delayed degenerative
diseases” by Olshansky and Ault in 1986.3 We are now seeing yet another
substitution of mortality due to later onset neurodegenerative disorders, such
as dementia and Parkinson’s disease (PD).4;5 It is predicted that the neurode-
generative disorders will gradually replace the systemic degenerative disorders
as the major causes of both death and morbidity in the 21st century.

Population Ageing

Population ageing is the product of three factors: birth rate, infant mortality
and life span. During the 19th century, and the first half of the 20th, these
three demographic factors were largely determined by major external or
environmental assaults due to infectious diseases, malnutrition and trauma.
Reductions in these risk factors resulted in improvements in maternal and
child health and increased infant survival. This was followed by declining
fertility and a decreased birth rate leading towards zero population growth;
hence the almost instantaneous ageing of Western populations in the first half
of the 20th century. Population ageing is one area of human ageing that can-
not be claimed by the geneticists as their responsibility. So far it is primarily
environmental.
The History of Ageing Group in Cambridge examined birth and death
records in five parishes in England between 1541 and 1981 to produce an

1541 1751 1921 1981

Figure 1. The proportion of elderly in the English population, 1541-1981 (adapted

from Laslett,6).
AGEING, HUMAN LIFESPAN AND NEURODEGENERATIVE DISORDERS 13

accurate projection of the number of over –60s in the English population dur-
ing a period of 440 years6 (Fig. 1). For almost 400 of those 440 years the over
–60s fluctuated at around 8% of the population. A consistent rise above 8%
did not occur until the 1920s or a mere 80 years ago when rapid population
ageing in developed countries commenced.

The Epidemiologic Transition Theory

The epidemiologic transition theory 2,3,7 explains these population changes by

major shifts in health and disease patterns, and in recorded causes of death,
as societies change or “modernise” with resultant improvement in social,
economic and health factors.
The base or first stage of the epidemiologic transition, graphically described
by Omran2 as “The Age of Pestilence and Famine,” was characterised by very
high death rates due to pandemic infections, trauma, poverty and malnutri-
tion. The major killers during this long pre-industrial era were the ubiquitous
diarrhoeas, influenza, tuberculosis and pneumonia, as well as epidemics such
as bubonic plague and small pox. The major death toll occurred in infants and
young children, with low average life span and a low and static percentage of
older people in the population.
The next stage of the epidemiologic transition followed the scientific
revolution and the start of the industrial revolution in England and Europe
in the 18th and 19th centuries. Concomitant social and economic changes
gradually brought greater wealth, better nutrition, less crowding, better
education, better hygiene, healthier mothers and stronger infants despite the
social upheavals of industrialisation and urbanisation. Omran’s second epi-
demiologic stage, “The Age of Receding Pandemics,” with decreasing infant
mortality and an ongoing high birth rate, resulted in more infant survivors
and an overall younger population in England up to the mid-19th century.
Declining fertility then lead to progressive population ageing, with a shift in
causes of death to the later-onset systemic degenerative diseases (particularly
cardiovascular and lung disease) and a shift in mortality from the young to
the old. These shifts heralded Omran’s third epidemiologic stage “The Age

Table 1. The Epidemiologic Transition Theory (Western model).

• The Age of pestilence and famine

• The Age of receding pandemics

• The Age of degenerative diseases

• The Age of delayed degenerative diseases

Omran (1971)2 and Olshansky & Ault (1986)3

14 THE AGEING BRAIN

of Degenerative and Man-made Diseases”. In retrospect, this stage represents

a major achievement for the ageing survivors of an era of fatal infectious
diseases, rather than a “man-made epidemic” of modern life, as it has often
been painted.
At the time of publication of his general theory of epidemiologic transition
and mortality change in 1971, Omran and other demographers were predict-
ing that average human life span would not progress beyond 70 years, then
considered to be the biological as well as the biblical limit. Omran himself
believed his third stage of degenerative diseases would be the completion
of the epidemiologic transition. However mortality rates at older ages have
continued to decline and average life expectancy at birth has continued to
increase worldwide. These changes lead to the description of a fourth stage
of the epidemiologic transition “The Age of Delayed Degenerative Diseases”,
by Olshansky and Ault in 1986.3 This stage recognized the rapid decline in
mortality due to chronic systemic diseases from the 1960s, particularly a
decline in cardiovascular disease and stroke in developed countries. There was
a delay in the ages at which these potentially fatal systemic diseases tended
to kill, with rapid improvement in life expectancy concentrated among the
population at advanced ages; a phenomenon described as “the ageing of the
aged.” This ongoing decline in mortality has been attributed to new public
health measures, including changes in major risk factors for systemic degen-
erative diseases such as smoking, diet and exercise, as well as advances in
medical technology and drugs.3,8 However Olshansky3,9 has predicted that
increases in life expectancy due to the prevention or delay of the known
systemic degenerative diseases would not increase average life expectancy at
birth much beyond 85 years.

Lifespan and Compression of Morbidity

The epidemiologic transition theory has focused on mortality with only

implicit reference to morbidity, defined as the length and quality of survival
in the presence of age-associated disease or disability.7 Description of the
phenomenon of “ageing of the aged”, with recognition of significant increases
in average life span beyond seven decades, brought increased attention to
the concepts of “healthy ageing” or “successful ageing” with emphasis on
the duration of disability-free survival, rather than longevity per se. James
Fries10 outlined his theory of “Compression of Morbidity” in 1980. Based on
a human life span of around 85 years, Fries predicted that chronic systemic
disease, and consequent disability, would be delayed and compressed to the
end of life by ongoing changes in life style and risk factors such as reduc-
tion in smoking, improved diet and more exercise. Recent data support this
association.11
However, with further increases in human life span, it remains possible
that morbidity is simply being delayed to later decades of life rather than
AGEING, HUMAN LIFESPAN AND NEURODEGENERATIVE DISORDERS 15

Figure 2. The theory of “compression of morbidity” as outlined by James Fries.10

Fries predicted that chronic systemic disease and associated disability
would be delayed and compressed to the end of life by ongoing changes in
lifestyle and risk factors.

compressed, i.e. from the “young-old” to the “old-old” or from the 70s to the
80s and 90s. A shift to a new epidemiologic transition of later-onset neuro-
degenerative disorders would result in additional causes of morbidity as well
as mortality at advanced ages. While morbidity related to chronic systemic
diseases appears to be declining, or being compressed to the end of life, the
morbidity related to neurodegenerative disease, particularly the dementias, is
increasing with advanced old age. Health related factors (reduction in smok-
ing, improved diet and more exercise, etc.) reducing mortality and morbidity
due to chronic systemic diseases have not been demonstrated to be protective
against the chronic neurodegenerative diseases. Other protective factors for
these diseases may be found, however, particularly those related to early brain
growth and development, and to intellectual ability and education in early
life.
Average human life span in developed countries is now approaching the
“natural limit” of 85 years described by Fries10 and predicted by Olshansky9
on the basis of possible cures for the major (systemic) degenerative diseases:
cardiovascular diseases, lung diseases, and cancer. More extreme longevity
remains possible, with average human life span going beyond the predicted
85 years and up to 100 or more years, but only if new causes of mortality
decline are determined and modified, additional to those responsible for the
16 THE AGEING BRAIN

decline in later onset systemic diseases, or new factors determining longevity

are identified. Important factors for further increases in longevity are likely
to be those related to brain development and to lifelong improvements in
cognitive and behavioural capacity.
The worldwide nature of rapid population ageing12 and the timing of
improvements in old age survival from the 1950s (before major advances in
the “new” public health) suggest that general social and biodemographic fac-
tors, as well as health factors, are producing the improvement in the survival
of the “old-old”. Vaupel’s group13 have demonstrated a substantial increase
in human survival, commencing in the 1950s, with mortality data showing
unpredicted and unexplained decline in mortality in those over 80 years of age
accompanied by a marked rise in absolute numbers of the “old-old”. Their
data indicate that the population of centenarians in developed countries has
doubled every decade since 1960, mostly as a result of increases in survival
after 80 years of age. This improvement in late-life survival is primarily non-
genetic. It is largely determined by early life factors and experiences, which
influence late life survival attributes, rather than by current conditions or risk
factors operating in late life. Individual life span is seen as a product of inter-
nal (including genetic) defence mechanisms or survival attributes and external
assaults on those defence mechanisms. External assaults, such as childhood
infections, malnutrition or trauma, may overwhelm internal defences and lead
to rapid or early death, as was common in the 19th century.
However, the survivors of these external assaults in early life may improve
their internal defence mechanisms (survival attributes) and lengthen their
subsequent life spans in old age. Vaupel’s group13 have also shown that death
rates decelerate with advanced age in multiple species: humans, medflies,
wasps, drosophila, nematodes and yeast cells. From the combined data, they
postulate that mortality decline with advancing age is a property of many
complex systems. It appears to be related to a cohort “survivor effect” trans-
mitted through individual fixed survival attributes, in an environment with
markedly reduced external assaults compared to previous centuries. This late
life “survivor effect” will apply to half the population in developed countries,
as average survival reaches 80 years of age.

Human Lifespan and the Brain

In terms of human life span, it is proposed that brain function responsible for
the human capacities for learning, cognition, insight and social knowledge,
is one determinant of longevity in human populations.13-15 Socio-economic
status, educational level, and mental ability or intelligence are closely linked.
A cohort effect of increasing fluid intelligence, as measured by psychometric
tests of verbal reasoning, spatial orientation and inductive reasoning, has been
demonstrated over the 20th century in data that span the period from 1889
to 1966.16 This cohort effect, which has been attributed to improvements in
AGEING, HUMAN LIFESPAN AND NEURODEGENERATIVE DISORDERS 17

education, parallels observed changes in both education and longevity over

the same period. Although no causal links have previously been suggested, it
is arguable that improvements in education and fluid intelligence are in part
responsible for increases in longevity. Socioeconomic status in childhood has
been associated with mortality from a number of illnesses17;18 and educa-
tional level contributes to differences in mortality.19;20 Higher mental ability
on test performance also correlates with better educational and occupational
outcomes.21
However, until recent longitudinal observations on Scottish school chil-
dren,22 there were few well-studied links between mental ability and mor-
tality. These included an Australian Vietnam Veteran study23 and the Can-
berra24 and Rotterdam25 studies on older populations. The larger Scottish
study was carried out on a cohort of 2792 school children in Aberdeen, with
mental ability assessed at 11 years of age (in 1932) and survival determined
65 years later on 80% of the sample. It concluded that childhood mental abil-
ity was a significant factor among the variables that predicted age at death
and hence longevity. The effects of IQ are difficult to separate from the effects
of social class and education. It can be argued, however, that better brain
development, whether it be in utero or in infancy and childhood, is likely
to be an important survival attribute in 21st century society with its greatly
reduced level of major external assaults and physical risk factors compared
to previous centuries. It can also be argued that average human life span is
likely to go beyond the 85 years which Olshansky has predicted on the basis
of projected reductions in mortality from chronic systemic diseases.9 The
important determinants of both mortality and morbidity in the “old-old”, in
the knowledge-based societies of the 21st century, are likely to be better brain
function on the one hand, and the neurodegenerative diseases associated with
brain ageing on the other.

The Neurodegenerative Diseases

The important late-onset neurodegenerative diseases, for the determination

of mortality and morbidity data in the older population, are dementia and
Parkinson’s disease. Despite their very high prevalence in the “old-old”, the
neurodegenerative diseases are, in general, poorly defined and diagnosed
compared to the common systemic degenerative diseases (heart disease,
stroke, chronic lung disease and cancer) and there is a high current level of
under-ascertainment of neurodegenerative disease mortality.5 The late-onset
neurodegenerative diseases include the dementias (Alzheimer’s disease [AD],
dementia with Lewy bodies [DLB] and fronto-temporal dementia [FTD]) as
well as Parkinson’s disease [PD] and motor-neuron disease [MND]). The
commonest cause of visual loss in older people, age-related macular degen-
eration (ARMD), can be classified as a neurodegenerative disease, as can the
almost universal age-related sensori-neural deafness. MND or amyotrophic
18 THE AGEING BRAIN

lateral sclerosis is the third commonest well-defined neurodegenerative dis-

ease of ageing. The term is also used for a host of less common familial and/
or sporadic neurological diseases of unknown cause, including progressive
supranuclear palsy (PSP), cortico-basal degeneration (CBD) and the spino-
cerebellar atrophies (SCA), many of which appear to be age-related.26
As a class, the neurodegenerative diseases are primary neuronal disorders,
i.e. not secondary to known vascular, malignant or toxic causes. Their defin-
ing feature is selective neuronal loss in a pattern that tends to be specific to
each disease. Many neurodegenerative diseases (AD, PD, MND) manifest as
a more common late onset sporadic form, which increases exponentially in
incidence with advancing age over 70 years, and rare early onset dominantly
inherited forms of what appear to be the same disease process. A number of
neurodegenerative diseases are characterized by the accumulation, over many
decades, of abnormal gene products in the brain; these proteins have vari-
able associations with the selective patterns of neuronal loss observed in each
disease. They include the accumulation of β-amyloid and tau in Alzheimer’s
disease, forms of synuclein in PD and DLB, and forms of tau in FTD and
other less common neurodegenerative diseases (PSP, CBD). The role of these
proteins remains poorly understood in the pathogenesis of the specific dis-
eases and particularly their late onset forms. Detailed study of the early onset
familial forms is, however, providing significant insights into the role of some
specific gene products. β-Amyloid in particular clearly plays an important
role in the commonest age-related neurodegenerative disease, Alzheimer’s
disease, and appears to have an additional role in brain ageing. Furthermore,
β-amyloid, in conjunction with the evolution of the apolipoprotein alleles in
humans, may have an association with basic evolutionary processes determin-
ing ageing and longevity.27

Neurodegenerative Diseases and Mortality

The first four stages of the epidemiologic transition have been defined by
mortality data using life expectancy and survival curves, combined with mor-
tality data on specific causes of death.2,3 The focus has been on deaths from
infectious diseases, and from the rapidly fatal systemic diseases, in particular
mortality data for heart disease, stroke, lung disease and cancer,3 which are
the commonest recorded causes of death in developed countries. They are
diseases that tend to have well defined fatal outcomes, and mortality data for
these disease categories is likely to be accurate. Clinical diagnosis is also likely
to be accurate for the other systemic diseases listed among the 10 common
causes of death in developed countries including endocrine, gastrointestinal
and genito-urinary causes. Age-standardised mortality, as well as morbidity,
for most of these systemic disease categories has been shown to be declining
over the 20th century in developed countries, including Australia, with the
most recent decline occurring in cancer deaths.28-30 The major exceptions
AGEING, HUMAN LIFESPAN AND NEURODEGENERATIVE DISORDERS 19

to this pattern of declining mortality from the Australian data are late-onset
neurological diseases (over 75 years) and later onset “mental health disor-
ders” (over 85 years).
While the quality of mortality data for both infectious and systemic degen-
erative diseases is relatively good, the same is not true for mortality data for
the neurodegenerative diseases, where under-ascertainment is a significant
problem.5,31 Although dementia is the commonest neurodegenerative disease,
its prevalence and incidence had not been well defined up until the last dec-
ades of the 20th century, particularly in the “old-old”. It is now clear that the
prevalence and incidence of dementia rise exponentially with age at least up
to 90 years.32,33 Alzheimer’s disease is the commonest dementia, and shows
a doubling of incidence every five years from 65 to 90 years of age.33 There
has been controversy as to whether dementia incidence plateaus off over 90
years of age, with two meta-analyses producing conflicting results.33,34 How-
ever a decline in AD incidence over 90 years, in men, is now supported by
the recent EURODEM project35 and in both men (over 93 years) and women
(over 97 years) by the Cache County study.36 Diagnosis of dementia during
life remains difficult in old age and a number of studies demonstrate the lack
of recognition of dementia in the community by family informant37 medical
practitioners38 and nurses.39 Death certificates tend to record the “acute”
systemic cause of death and mortality from death certificate data is estimated
to be as low as 15% for dementia.5 Finally, studies suggest that Vascular
dementia (VaD), a dementia of systemic cause, does not rise as rapidly with
age as AD40,41 and that mixed dementia (AD and VaD) is common.42,43The
inclusion of VaD in mortality and morbidity data for neurodegenerative dis-
eases is often difficult to avoid, but should not greatly distort the results.5
Parkinson’s disease (PD), the second commonest neurodegenerative dis-
ease, is also poorly defined and diagnosed in older people, in whom atypical
forms of the disease are more common.44 Until recently, idiopathic PD was
stated to decline in incidence with ageing over 75 years in studies based on
inappropriate methodology.45 It is now clear that its prevalence and incidence
continue to rise with advanced ageing.26,42,46 This rising age-related preva-
lence may not be well recorded in mortality data. Only 25% of decedents with
PD have the disease listed on their death certificates.5
Finally, in terms of the quality of mortality data, neurodegenerative dis-
orders are commonly mixed in the “old-old”.42,43 Furthermore multiple pre-
clinical syndromes commonly co-exist in older people and have been shown
to predict subsequent dementia;47 these include cognitive or memory impair-
ment (not reaching criteria for AD), motor slowing (not reaching criteria
for PD) and evidence of vasculopathy. Because of multiple pathology in the
“old-old”, the neurodegenerative disorders outlined commonly present as
multi-factorial “Geriatric Syndromes”, rather than as specific neurological
diseases amenable to specific diagnoses on death certificates. Many of the
“Geriatric Syndromes” have a high mortality rate including: “immobility”
with underlying parkinsonism and dementia; “instability and falls” with
20 THE AGEING BRAIN

underlying impairments of balance gait and vision; “delirium” with underly-

ing frontal system impairments and dementia; and “aspiration pneumonia”
due to underlying brain and oesophageal-motility disorders. However, the
underlying causal diagnoses rarely appear on death certificates.
Despite these potential problems with certification of deaths due to neu-
rodegenerative diseases, a number of recent studies have shown increasing
mortality from the three major neurodegenerative diseases: dementia,5,31,48
Parkinson’s disease49,50 and MND.51 Few studies have been able to compare
mortality data for these neurodegenerative diseases with mortality data for
common systemic diseases. Lilienfeld and Perl5 used US Census Bureau popu-
lation estimates to project the annual death rate from three neurodegenerative
diseases (dementia, PD and MND) and from six comparison systemic dis-
eases (liver cirrhosis, colon cancer, lung cancer, cancer of the female breast,
multiple sclerosis, and malignant melanoma) over the period between 1990
and 2040. The US National Center for Health Statistics routinely collects
individual death certificates for all US residents. To determine death rates
they used data for deaths in which the underlying cause was dementia, PD
or MND (and the six comparison diseases) for the years 1985–1988. Assum-
ing that the US disease-age-gender-race-specific death rates for these years
remained constant over the period between 1990 and 2040, they found that
neurodegenerative disease mortality increased by 119–231%, depending on
the population model used. For the “middle” population growth model the
increase was 166%, with the major component being deaths due to demen-
tia. The increases in mortality for the six comparison diseases ranged from
52% (multiple sclerosis) to 130% (colon cancer). A number of factors make
it likely that these projections for neurodegenerative disease mortality are
underestimates including under-ascertainment on death certificates (for the
reasons outlined above) and the conservative nature of the US Census Bureau
estimates of population ageing. Furthermore the comparison with cancer
deaths is with a category of systemic disease in which mortality is either still
rising or static or showing the slowest falls, in comparison with other com-
mon systemic diseases, such as cardiovascular and lung disease, in which
mortality is declining rapidly.28;30
Based on this review of the limited mortality, life expectancy and survival
data, deaths from most systemic degenerative diseases continue to decline and
are being replaced by deaths from even later onset neurodegenerative diseases,
as part of a new disease transition.

Neurodegenerative Diseases and Morbidity

Overall, the quality of data to examine and compare morbidity and disability
by disease cause across populations, has been less accurate than mortality
data, with few reliable data on morbidity available.7 However it is increas-
ingly important to define and measure morbidity, taking into account the
AGEING, HUMAN LIFESPAN AND NEURODEGENERATIVE DISORDERS 21

delayed onset, slower course and reduced mortality associated with the
chronic systemic diseases over the past 50 years, the concomitant rise of the
age-related neurodegenerative diseases, and the current controversies about
compression of morbidity. This is particularly the case in comparing the
burden of chronic systemic disease with that of chronic neurodegenerative
disease, as it is the latter that is likely to rise in the 21st century with further
population ageing.
Many studies have shown rising age-related incidence, prevalence and
morbidity in individual neurodegenerative disorders, such as dementia32,33,52
and PD.26,46,50 More general population based studies using Australian
Bureau of Statistics (ABS) or US Census Data53,54 commonly rely on self-
report instruments to identify diseases and compare causes of disability, and
have not shown this trend. However such instruments may not be sensitive to
the impact of neurodegenerative diseases on disability, since cognition and/or
insight are commonly impaired.55 This has been confirmed for the ABS dis-
ability instrument in a study comparing different disability measures, given
to both respondents and informants.56
Few studies have used detailed clinical assessments in the field or compared
incidence, prevalence or morbidity data, or disability rates, due to neurode-
generative diseases as a group with those rates due to the common systemic
diseases. The Kilsyth Study in Scotland,40,57,58 completed in early 1970,
examined the prevalence of major chronic disorders in the elderly and com-
pared disability and dependence due to systemic and neurological causes. The
study involved three community-living random samples, comprising 808 peo-
ple, 65 years and over, examined by physicians experienced in geriatric medi-
cine.58 It demonstrated that the prevalence of disability for IADL (defined as
the inability to live at home without domestic help) increased from 12% at
65–69 years to over 80% at the age of 85 years. It showed that neurological
disorders — dementia, balance/gait disorder, stroke and parkinsonism in that
order — were the commonest cause of disability in 48% of subjects. Neu-
rological and functional psychiatric disorders together contributed to 70%
of disability compared with cardio-respiratory (38%), joint disease (24%)
obesity (16%) and vision (11%). Neurological disorders, in 93% of cases
(particularly dementia in 77% of cases), were by far the greatest contributor
to the more severe category of dependence in ADL (defined as impairment
in personal care) followed by joint disease (30%), cardio-respiratory (18%),
vision (15%) and obesity (11%).
Subsequent studies suggest that chronic systemic diseases have been delayed
and compressed over the decades since the Kilsyth Study was completed in
1970,3,9-11 and the morbidity of specific neurodegenerative disorders, such as
dementia32,52 and PD,26,46 has risen exponentially with advanced ageing of
the population. The only known study in the subsequent two decades com-
paring prevalence and disability data in older people, for a range of common
chronic systemic disorders with neurodegenerative disorders, is the Sydney
Older Persons Study (1991 to 2002). This longitudinal study comprised two
22 THE AGEING BRAIN

Figure 3. The age distribution, by gender, in the Sydney Older Persons Study, Wave
1 (Waite et al., 1997)42.

random samples of 647 community-living people aged 75 to 98 years, with

equal numbers of men and women. 59,42,43,47,56,60-62 The study provides
prevalence data, on the same neurological and systemic disorders as the Kil-
syth Study, in 522 subjects who agreed to a detailed medical, neurological,
psychometric and disability assessment by a physician experienced in geriatric
medicine.42
The 392 survivors examined at Wave 2 of the study, three years later, pro-
vided incidence data. The numbers assessed were modest, but the prevalence
and incidence of both systemic and neurodegenerative disorders at this age
range is very high, enabling examination of the data to look at the concept of
a new epidemiologic transition in terms of causes of morbidity in an ageing
population.
The six major chronic systemic disorders measured included five shown to
be the causes of disability from the Kilsyth study (heart disease, stroke, respi-
ratory disease, arthritis and obesity), with the addition of peripheral vascular
disease. The six neurodegenerative disorders measured included dementia, Par-
kinson’s disease, visual impairment, and a disorder of gait and balance, which
had been identified as significant neurodegenerative causes of disability in the
Kilsyth Study. Gait and balance disorder was further divided into two clinical
components measured as motor slowing (gait slowing in subjects not reaching
standard criteria for Parkinson’s disease) and gait ataxia (impaired heel-toe
gait performance). Mild cognitive impairment (in subjects not reaching stand-
ard criteria for dementia) was included as the sixth neurodegenerative disorder
measured. The data presented in Figures 4 and 5 are given in three- year age
bands using smoothed estimates from a logistic regression model.
AGEING, HUMAN LIFESPAN AND NEURODEGENERATIVE DISORDERS 23

Figure 4. The prevalence of chronic systemic diseases, in three year age bands com-
bined for males and females, in the Sydney Older Persons Study, Wave 1
using smoothed estimates from a logistic regression model (Waite et al.,
1997)42. (N=522; *p <0.05; ** p<0.01.)

The prevalence data for the chronic systemic diseases were examined for
men and women in 20 percentile age bands between 75 and 84 years.42 The
data showed the expected high rates of systemic disease in this very old popu-
lation, ranging from 70% for arthritis and 46% for heart disease to 20% for
chronic lung disease, 16% for stroke, 14% for obesity and 11% for periph-
eral vascular disease. The expected male-to-female differences in prevalence
were observed, particularly in chronic lung disease (29% to 12%), peripheral
vascular disease (14% to 8%) and obesity (8% to 20%). However the sys-
temic diseases did not increase in prevalence with advancing age between 75
and 90 years, most showing a trend to decline which was significant only for
chronic lung disease. The three-year incidence data confirmed this picture and
showed the same trend towards static or declining incidence for the systemic
diseases with advancing age over 75 years.
The neurodegenerative disorders, examined in the same 20 percentile
age bands,42 show similarly high overall prevalence rates, however without
marked gender differences. These ranged from a prevalence of 50% for gait
ataxia and 43% for visual impairment to 38% for cognitive impairment,
19% for gait slowing, 17% for dementia and 5% for Parkinson’s disease.
All six neurodegenerative disorders examined showed a marked and highly
significant increase in prevalence with advancing age from 75 to 93 years.
24 THE AGEING BRAIN

Figure 5. The prevalence of neurodegenerative disorders, in three year age bands

combined for males and females, in the Sydney Older Persons Study, Wave
1 using smoothed estimates from a logistic regression model (Waite et al.,
1997)42. (N=522; *p <0.05; ** p<0.01.)

The three-year incidence data confirmed this picture, showing rapidly increas-
ing incidence with advancing age over 75 years for all six neurodegenerative
disorders.
A detailed analysis of disability data was carried out in terms of personal
care (ADL), domestic care (IADL) and mobility for all clinical diagnoses using
the Kilsyth Disability Scale;57 disability rates were compared between the sys-
temic and neurodegenerative disorders.56 Neurodegenerative disorders were
prominent contributors to all three areas of disability; they were the major
contributors to both ADL and IADL disability, as measured by the Kilsyth Dis-
ability Scale, given independently to both the subjects and to their informants.
In contrast to the Kilsyth Disability Scale the traditional ABS disability scale
completed by the subject identified arthritis as having a large impact on dis-
ability but was not sensitive to the impact of the neurodegenerative disorders.
Using the Kilsyth Instrument,57 disability was estimated for each of the
disorders identified. A proportion of disability was then attributed to either
the group of six systemic degenerative disorders or the group of six neuro-
degenerative disorders for two waves of the Sydney Older Persons Study.
As shown in Figure 6, around 70% of all disability was attributable to the
neurodegenerative disorders.
Based on this review including the limited morbidity data available, together
with the previously discussed mortality, life expectancy and survival data, dis-
AGEING, HUMAN LIFESPAN AND NEURODEGENERATIVE DISORDERS 25

Figure 6. Proportions of disablity attributable to neurodegenerative and systemic

disorders in in the Sydney Older Persons Study, Waves and 2 (Waite et al.,
1997)56.

ability due to the chronic systemic diseases is being reduced or delayed, as pre-
dicted by Fries10 and Olshansky.3 However systemic disease related disability
is being replaced by a new wave of disability due to neurodegenerative disor-
ders, which are increasing exponentially in the most rapidly growing sector of
the population, the “old-old”, forming a new disease transition.

A Fifth Epidemiologic Transition: The Age of Neurodegenerative Disorders

Rapid population ageing is now a worldwide phenomenon. The trend to

“ageing of the aged”, with the biggest increases in the population occurring
in those 80 years and over, is most immediately apparent and relevant to the
countries of the European Union, Australia, the USA and Japan and, a little
further down the track, to the rapidly ageing populations of Eastern Europe
and Southeast Asia. In the developed world, a majority of those who survive
to 80 years and over will go on to develop neurodegenerative diseases before
death. In contrast, while the concept of a “delayed transition” in poorer
countries is of great global importance for other reasons,7 age-related neu-
rodegenerative disorders are not likely to be one of the important issues for
these societies in the near future.
26 THE AGEING BRAIN

The conclusion, from this review of the literature and comparison of

mortality, morbidity and disability data and trends, is that the late-onset
neurodegenerative disorders will cause increasing morbidity towards the end
of the increasing human life span in developed countries. Further they are
gradually replacing the fatal systemic diseases, such as heart disease, stroke
and respiratory disease, and ultimately cancer, as the major causes of death.
This process parallels the exponential increase in the numbers of those sur-
viving into their 80s and 90s. This is the age at which the neurodegenerative
diseases show their greatest increase in prevalence and incidence, and the age
at which the chronic systemic diseases show static or falling rates, in part due
to the success of current public health measures.
While a vast array of social, economic, environmental and demographic
factors determine the course of population change, it is clear that population
ageing and human lifespan are closely related to risk and preventive factors
for disease. The theory of Epidemiologic Transition, as outlined here, is an
attempt to explain the origin and complexity of the changing mortality and
morbidity patterns towards neurodegenerative diseases in an ageing popula-
tion, in order to emphasise the enormous impact these diseases will have on
future public health programs and on the health care industry, including acute
hospital care and community health services, in the coming decades.
The success of public health measures over the last 50 years, with changes
in risk factors such as smoking, diet, exercise and lifestyle factors, has indeed
lead to a delayed onset of systemic degenerative diseases and a probable
“compression of morbidity” from those diseases towards the end of life, as
Fries predicted in 1980.10 However there is as yet no evidence that attention
to risk factors identified for systemic degenerative diseases will reduce the
incidence, or delay the onset, of the primary neuronal disorders underlying
such neurodegenerative diseases as AD, PD, DLB or MND. The neuronal
systems involved in these disorders do not appear, on current evidence, to
be susceptible to the same life style or risk factors that devastate other body
systems. With further increases in human life span, it remains possible, and
indeed likely, that morbidity will simply be delayed to later in the life span,
i.e. from the “young-old” to the “old-old” or from the 70s to the 80s and 90s,
rather than compressed; unless new risk factors to delay the primary neuronal
disorders are identified.
The important question for ageing research is not only “is this degenera-
tive process ageing or disease?” but, more significantly, “can this degenera-
tive process be modified, prevented or delayed, without significant risk, by
manipulation of environmental and/or genetic risk factors?”. The aim of
ageing research remains one of compression of morbidity, in this case from
the neurodegenerative disorders, towards the end of life and the prolonga-
tion of the period of healthy non-disabled life. This aim may, or may not, be
consonant with increased longevity. The challenge facing ageing research is to
seek new and modifiable risk factors to delay the onset of neurodegenerative
disorders which are reducing quality of life in advanced old age. While the
AGEING, HUMAN LIFESPAN AND NEURODEGENERATIVE DISORDERS 27

study of genetic risk factors has dominated recent research publication in

AD, there remains good evidence for associated environmental factors, either
causal, or as modifiers of timing of disease onset or rate of progression, both
of which mechanisms might alter prevalence or morbidity.36,63,64 There is
recent evidence that the commonest neurodegenerative disorder, AD, may
decline in incidence with advancing age over 90 years35,36 and factors such
as educational level and brain size may modify the onset of AD65–68. There
are major current research endeavours examining potential genetic and envi-
ronmental modifiers for the progression of AD.63 Finally, both Omran2 and
Olshansky3 were inaccurate in their prediction that we had come to the end of
the Epidemiologic Transition, in 1971 and 1986, respectively. There is good
reason to believe that we will have yet another transition, that of delayed
neurodegenerative disorders, early in the 21st century.

Acknowledgements

This work has been supported by a series of National Health and Medical
Research Council Grants over the past decade and by Grants from the Depart-
ment of Veteran Affairs, the Ageing and Alzheimer’s Research Foundation and
the Centre for Education and Research on Ageing of the University of Sydney
and by the resources of the Prince of Wales Medical Research Institute. I am
indebted to the support of Professor Francis Caird for the work carried out
in the Kilsyth Study (1970-71). I am also indebted to my PhD students for
much of the work carried out in the Sydney Older Persons Study (1982-2002),
including Dr Louise Waite, Dr Olivier Piguet, Dr Hayley Bennett, Dr Wayne
Reid and Dr Tanya Lye; to my colleagues in that study including Associate
Professor Dave Grayson, Dr Helen Creasey, Dr William Brooks, Dr Glenda
Halliday and Dr Jillian Kril; and to the administrative support for that study
from Sandra Forster, Jill Groth and Jan Koh.

References

1. Eberstadt N. The population implosion. Foreign Policy. 2001; 123:42–53.

2. Omran AR. The epidemiologic transition. A theory of the epidemiology of popu-
lation change. Milbank Q. 1971; 49:509–538.
3. Olshansky SJ, Ault AB. The fourth stage of the epidemiologic transition: The age
of the delayed degenerative diseases. Milbank Q. 1986; 64:355–391.
4. Broe GA, Creasey H. Brain ageing and neurodegenerative disease: A major public
health issue of the the twenty first century. Perspect Hum Biol. 1995; 1:53–58.
5. Lilienfeld DE, Perl DP. Projected neurodegenerative disease mortality in the
United States, 1990-2040. Neuroepidemiology. 1993; 12:219–228.
6. Laslett P. The significance of the past in the study of ageing. Ageing Soc. 1984;
4:379–389.
7. Phillips DR. Problems and potential of researching epidemiological transition:
examples from Southeast Asia. Soc Sci Med. 1991; 33:395–404.
28 THE AGEING BRAIN

8. Fries JF. Can preventive gerontology be on the way? Am J Public Health. 1997;
87:1591–1593.
9. Olshansky SJ, Carnes BA, Cassel C. In search of Methuselah: estimating the upper
limits to human longevity. Science. 1990; 250:634–640.
10. Fries JF. Aging, natural death, and the compression of morbidity. N Engl J Med.
1980; 303:130–135.
11. Vita AJ, Terry RB, Hubert HB, Fries JF. Aging, health risks, and cumulative dis-
ability. N Engl J Med. 1998; 338:1035–1041.
12. Laslett P. Interpreting the demographic changes. Phil Trans R Soc Lond B Biol
Sci. 1997; 352:1805–1809.
13. Vaupel JW, Carey JR, Christensen K, Johnson TE, Yashin AI, Holm NV,
Iachine IA, Kannisto V, Khazaeli AA, Liedo P, Longo VD, Zeng Y, Manton KG,
Curtsinger JW. Biodemographic trajectories of longevity. Science. 1998; 280:
855–860.
14. Kaplan H, Hill K, Lancaster J, Hurtado AM. A theory of human life history
evolution: Diet, intelligence, and longevity. Evol Anthropol. 2000; 9:156–185.
15. Rapoport SI. How did the human brain evolve? A proposal based on new evidence
from in vivo brain imaging during attention and ideation. Brain Res Bull. 1999;
50:149–165.
16. Schaie KW. Intellectual development in adulthood. In Birren JE, Schaie KW, edi-
tors. Handbook of the psychology of aging. San Diego: Academic Press, 1996:
266–286.
17. Joseph KS, Kramer MS. Review of the evidence on fetal and early childhood
antecedents of adult chronic disease. Epidemiol Rev. 1996; 18:158–174.
18. Smith GD, Hart C, Blane D, Hole D. Adverse socioeconomic conditions in child-
hood and cause specific adult mortality: prospective observational study. Brit
Med J. 1998; 316:1631–1635.
19. Doornbos G, Kromhout D. Educational level and mortality in a 32-year follow-
up study of 18-year-old men in The Netherlands. Int J Epidemiol. 1990; 19:
374–379.
20. Kunst AE, Mackenbach JP. The size of mortality differences associated with
educational level in nine industrialized countries. Am J Public Health. 1994; 84:
932–937.
21. Neisser U, Boodoo G, Bouchard TJJ, Boykin AW, Brody N, Ceci SJ, Halpern
DF, Loehlin JC, Perloff R, Sternberg RJ, Urbina S. Intelligence: Knowns and
unknowns. Am Psychol. 1996; 51:77–101.
22. Whalley LJ, Deary IJ. Longitudinal cohort study of childhood IQ and survival up
to age 76. Brit Med J. 2001; 322:819.
23. O’Toole BI, Stankov L. Ultimate validity of psychological tests. Pers Indiv Differ.
1992; 13:699–716.
24. Korten AE, Jorm AF, Jiao Z, Letenneur L, Jacomb PA, Henderson AS, Chris-
tensen H, Rodgers B. Health, cognitive, and psychosocial factors as predictors of
mortality in an elderly community sample. J Epidemiol Commun H. 1999; 53:
83–88.
25. Deeg DJ, Hofman A, van Zonneveld RJ. The association between change in
cognitive function and longevity in Dutch elderly. Am J Epidemiol. 1990; 132:
973–982.
26. Tanner CM, Aston DA. Epidemiology of Parkinson’s disease and akinetic syn-
dromes. Curr Opin Neurol. 2000; 13:427–430.
27. Finch CE, Sapolsky RM. The evolution of Alzheimer disease, the reproductive
schedule, and apoE isoforms. Neurobiol Aging. 1999; 20:407–428.
28. Beaglehole R. International trends in coronary heart disease mortality, morbidity,
and risk factors. Epidemiol Rev. 1990; 12:1–15.
AGEING, HUMAN LIFESPAN AND NEURODEGENERATIVE DISORDERS 29

29. Australian Bureau of Statistics. 1989–1990 national health survey. Health status
indicators. Catalogue No.4370.0. Canberra: ABS, 1992.
30. d’Espaignet ET, van Ommeren M, Taylor F, Briscoe N, Pentony P. Trends in
Australian mortality: 1921–1988. Canberra: Australian Institute of Health and
Welfare, 1991.
31. Baldereschi M, Di Carlo A, Maggi S, Grigoletto F, Scarlato G, Amaducci L, Inzi-
tari D. Dementia is a major predictor of death among the Italian elderly. ILSA
Working Group. Italian Longitudinal Study on Aging. Neurology. 1999; 52:
709–713.
32. Jorm AF, Korten AE, Henderson AS. The prevalence of dementia: a quantitative
integration of the literature. Acta Psychiat Scand. 1987; 76:465–479.
33. Jorm AF, Jolley D. The incidence of dementia: a meta-analysis. Neurology. 1998;
51:728–733.
34. Gao S, Hendrie HC, Hall KS, Hui S. The relationships between age, sex, and the
incidence of dementia and Alzheimer disease: a meta-analysis. Arch Gen Psy-
chiat.1998; 55:809–815.
35. Andersen K, Launer LJ, Dewey ME, Letenneur L, Ott A, Copeland JRM, Dar-
tigues JF, Kragh-Sorensen P, Baldereschi M, Brayne C, Lobo A, Martinez-Lage
JM, Stijnen T, Hofman A. Gender differences in the incidence of AD and vascular
dementia: The EURODEM Studies. The EURODEM Incidence Research Group.
Neurology. 1999; 53:1992–7.
36. Miech RA, Breitner JC, Zandi PP, Khachaturian AS, Anthony JC, Mayer L. Inci-
dence of AD may decline in the early 90s for men, later for women: The Cache
County study. Neurology. 2002; 58:209–218.
37. Ross GW, Abbott RD, Petrovitch H, Masaki KH, Murdaugh C, Trockman C,
Curb JD, White LR. Frequency and characteristics of silent dementia among eld-
erly Japanese-American men. The Honolulu-Asia Aging Study. J Am Med Assoc.
1997; 277:800–805.
38. Callahan CM, Hendrie HC, Tierney WM. Documentation and evaluation of cog-
nitive impairment in elderly primary care patients. Ann Intern Med. 1995; 122:
422–429.
39. Sorensen L, Foldspang A, Gulmann NC, Munk-Jorgensen P. Assessment of
dementia in nursing home residents by nurses and assistants: criteria validity and
determinants. Int J Geriatr Psych. 2001; 16:615–621.
40. Broe GA, Akhtar AJ, Andrews GR, Caird FI, Gilmore AJ, McLennan WJ. Neuro-
logical disorders in the elderly at home. J Neurol Neurosurg Psychiat. 1976; 39:
361–366.
41. Brayne C, Gill C, Huppert FA, Barkley C, Gehlhaar E, Girling DM, O’Connor
DW, Paykel ES. Incidence of clinically diagnosed subtypes of dementia in an
elderly population. Cambridge Project for Later Life. Brit J Psychiat,. 1995; 167:
255–262.
42. Waite LM, Broe GA, Creasey H, Grayson DA, Cullen JS, O’Toole B, Edelbrock
D, Dobson M. Neurodegenerative and other chronic disorders among people aged
75 years and over in the community. Med J Australia. 1997; 167:429–432.
43. Waite LM, Broe GA, Grayson DA, Creasey H. The incidence of dementia in an
Australian community population: the Sydney Older Persons Study. Int J Geriatr
Psych. 2001; 16:680–689.
44. Broe GA. Parkinson’s disease and related disorders. In: Grimley Evans J, editor.
Oxford texbook in geriatric medicine. Oxford: Oxford University Press, 1992:
546–557.
45. Koller W, O’Hara R, Weiner W, Lang A, Nutt J, Agid Y, Bonnet AM, Jankovic J.
Relationship of aging to Parkinson’s disease. In: Yahr MD, Bergmann KJ, editors.
Parkinson’s disease. New York: Raven Press, 1986: 317–321.
30 THE AGEING BRAIN

46. Baldereschi M, Di Carlo A, Rocca WA, Vanni P, Maggi S, Perissinotto E, Grigo-

letto F, Amaducci L, Inzitari D. Parkinson’s disease and parkinsonism in a lon-
gitudinal study: two-fold higher incidence in men. ILSA Working Group. Italian
Longitudinal Study on Aging. Neurology. 2000; 55:1358–1363.
47. Waite LM, Broe GA, Grayson DA, Creasey H. Preclinical syndromes predict
dementia: the Sydney older persons study. J Neurol Neurosur Ps. 2001; 71:
296–302.
48. Dewey ME, Saz P. Dementia, cognitive impairment and mortality in persons aged
65 and over living in the community: a systematic review of the literature. Int J
Geriatr Psych. 2001; 16:751–761.
49. Lilienfeld DE, Chan E, Ehland J, Godbold J, Landrigan PJ, Marsh G, Perl DP.
Two decades of increasing mortality from Parkinson’s disease among the US
elderly. Arch Neurol. 1990; 47:731–734.
50. Rajput AH, Offord KP, Beard CM, Kurland LT. Epidemiology of parkinsonism:
incidence, classification, and mortality. Ann Neurol. 1984; 16:278–282.
51. Lilienfeld DE, Chan E, Ehland J, Godbold J, Landrigan PJ, Marsh G, Perl DP.
Rising mortality from motoneuron disease in the USA, 1962-84. Lancet. 1989;
1:710–713.
52. Witthaus E, Ott A, Barendregt JJ, Breteler M, Bonneux L. Burden of mortality
and morbidity from dementia. Alzheimer Dis Assoc Disord. 1999; 13:176–181.
53. Manton KG, Stallard E, Corder L. Education-specific estimates of life expectancy
and age-specific disability in the U.S. elderly population: 1982 to 1991. J Aging
Health. 1997; 9:419–450.
54. Manton KG, Corder L, Stallard E. Chronic disability trends in elderly United
States populations: 1982-1994. Proc Nat Acad Sci USA. 1997; 94:2593–2598.
55. Ostbye T, Tyas S, McDowell I, Koval J. Reported activities of daily living: agree-
ment between elderly subjects with and without dementia and their caregivers.
Age Ageing. 1997; 26:99–106.
56. Waite LM, Broe GA, Grayson DA, Creasey H, Cullen JS, Casey B, Bennett HP,
Brooks WS. Clinical diagnoses and disability among community dwellers aged 75
and over. Australas J Ageing. 2001; 20:67–72.
57. Broe GA, Akhtar AJ. Assessment of activities of daily living in the elderly. In:
Israel L, Kozarevic D, Sartorius N, editors. Book of geriatric assessment. Basel:
Karge,r 1984: 241–242.
58. Akhtar AJ, Broe GA, Crombie A, McLean WM, Andrews GR, Caird FI. Disability
and dependence in the elderly at home. Age Ageing. 1973; 2:102–111.
59. Broe GA, Jorm AF, Creasey H, Casey B, Bennett H, Cullen J, Edelbrock D, Waite
L, Grayson D. Carer distress in the general population: results from the Sydney
Older Persons Study. Age Ageing. 1999; 28:307–311.
60. Broe GA, Creasey H, Jorm AF, Bennett HP, Casey B, Waite LM, Grayson DA,
Cullen J. Health habits and risk of cognitive impairment and dementia in old age:
a prospective study on the effects of exercise, smoking and alcohol consumption.
Aust NZ J Publ Health. 1998; 22:621–623.
61. Creasey H, Waite LM, Grayson DA, Bennett HP, Dent O, Broe GA. The impact
of the neurodegenerative disorders on ageing: An overview of the Sydney Older
Persons Study. Australas J Ageing. 2001; 20:11–16.
62. Waite LM, Broe GA, Creasey H, Grayson D, Edelbrock D, O’Toole B. Neurologi-
cal signs, aging, and the neurodegenerative syndromes. Arch Neurol. 1996; 53:
498–502.
63. Breitner JC. The end of Alzheimer’s disease? Int J Geriatr Psych. 1999; 14:
577–586.
64. Amaducci L, Lippi A. Risk factors for Alzheimer’s disease. Editorial. Int J Geriatr
Psych. 1997; 7:383–388.
AGEING, HUMAN LIFESPAN AND NEURODEGENERATIVE DISORDERS 31

65. Graves AB, Mortimer JA, Larson EB, Wenzlow A, Bowen JD, McCormick WC.
Head circumference as a measure of cognitive reserve. Association with severity
of impairment in Alzheimer’s disease. Brit J Psychiat. 1996; 169:86–92.
66. Snowdon DA, Kemper SJ, Mortimer JA, Greiner LH, Wekstein DR, Markes-
bery WR. Linguistic ability in early life and cognitive function and Alzheimer’s
disease in late life. Findings from the Nun Study. J Am Med Assoc. 1996; 275:
528–532.
67. Snowdon DA. Aging and Alzheimer’s disease: lessons from the Nun Study. Ger-
ontologist. 1997; 37:150–156.
68. Jorm AF, Creasey H, Broe GA, Sulway MR, Kos SC, Dent OF. The advantage of
being broad-minded: Brain diameter and neuropsychological test performance in
elderly war veterans. Pers Indiv Differ. 1997; 23:371–3777.
 SECTION II

CHARACTERISTICS
OF THE
AGEING BRAIN
Chapter 3

STRUCTURAL CHANGES
IN THE AGEING HUMAN
BRAIN
Jillian J. Kril

Introduction

The link between ageing and disease in the brain is strong and in many
instances the increased prevalence of brain disease in the elderly is taken as
evidence that the disease is as a result of increasing age. Indeed, it has been
suggested that there is a continuum between age-related pathology and dis-
eases such as Alzheimer’s disease (AD), and that a decline in brain function
and an increase in brain pathology are inevitable consequences of ageing.
However, these assertions are, in many cases, unfounded and there is increas-
ing evidence that brain pathology is not synonymous with brain ageing.
There are a number of obstacles to being able to definitively separate
ageing from disease. Some of these can be overcome with methodological
improvements to research studies or with advances in technology, while oth-
ers are inherent to the study of human disease. In addition, there are a number
of theoretical issues to be considered when discussing the concept of brain
ageing.

1. There is convincing evidence of a long “prodromal” or “preclinical”

period for diseases such as AD and, unless there is rigorous clinical and
pathological scrutiny of subjects for inclusion in studies of ageing, the
results of these studies will be biased towards the finding of an age-related
decline. Carefully controlled studies have demonstrated both functional
and pathological deficits in individuals who do not meet diagnostic cri-
teria for AD. For example, individuals who go on to develop dementia
36 THE AGEING BRAIN

may have a long period with stable deficits which precedes the disease
diagnosis.1,2 In addition, it has been shown that there is marked neuronal
loss from some brain region in patients with equivocal or early disease (i.e.
Clinical Dementia Rating = 0.53)4 suggesting that neuronal loss predates
the onset of symptoms. Taken together, these studies highlight the need
for: (i) the continual refinement and improvement of diagnostic tools to
identify early disease not only for dementia but for other diseases common
in the elderly, and (ii) post-mortem verification of cases used in functional
studies of ageing and disease to confirm diagnosis, exclude co-existing
diagnoses (e.g. other causes of dementia), and perform clinicopathological
correlations.
2. Increasing age is a risk factor for most neurodegenerative diseases as well as
many systemic disorders which also affect brain function. While this does
not mean that age and disease are inseparable, it does pose the question of
what is meant by “normal”. On one hand, normal may be interpreted as
meaning representative of most of the population, while in another sense
it may mean free from abnormality. In the case of brain ageing studies, if
these are performed on subjects which have been highly selected to exclude
the presence of all pathology, the results obtained may be artificial in that
they represent a small subset of aged subjects rather than being indicative
of the majority of older people. Conversely, if an unselected group of sub-
jects is studied, one runs the risk of falsely attributing the consequences of
some common, age-related disease to ageing per se. Mrak et al.5 state that
to be regarded as being due to the ageing process, changes must be both
universal and intrinsic. While both these requirements appear plausible, it
is difficult to see, given the enormous genetic variability in humans, how
universal can be assessed. For example, there is evidence that the relation-
ship between increasing age and increasing disease does not hold for those
over 95 years (the oldest old). A number of studies have demonstrated that
these subjects are healthier than many people in their eighties6 suggesting
resistance or delayed sensitivity in some individuals. Other confounding
factors such as differential survival rates will also have an impact on the
results of studies on ageing. Such constraints must be considered in the
design and interpretation of any study of ageing.
3. A true longitudinal pathological study cannot be performed. In rare cases
the brains of individuals are sampled on more than one occasion,7–10 and
while these offer the advantage of investigations in the same individual
over time, exactly the same tissue cannot be evaluated. In addition, there
are potential confounding issues associated with the patient having under-
gone a surgical procedure and the tissue reactions to this. Advances in
technology, especially neuroimaging, have allowed for better identifica-
tion of brain structures during life, but their resolution is still at the mil-
limeter level, which does not allow for accurate identification of cellular
populations in most cases, and few studies have been performed following
subjects to autopsy. This mean that the majority of pathological studies
STRUCTURAL CHANGES 37

are restricted to group comparisons and are therefore subject to the limita-
tions such cohort studies impose.

Brain Atrophy

A decrease in brain size with increasing age is widely accepted in medi-

cine.11,12 It is not uncommon for radiological or pathological reports to
dismiss the presence of cerebral atrophy as “consistent with the subject’s
age”, a practice which has perpetuated the concept of inevitable senescence.
While there is good evidence that the brain size of older individuals is, on
average, smaller than that of younger individuals, this represents a cohort
effect rather than true atrophy. Several studies have demonstrated a steady
increase in brain size for both men and women during the past century.11,13,14
Dekaban and Sadowsky13 reviewed the brain weights of over 4,500 persons
of all ages who were free from neurological disease and showed an 11% dif-
ference between young and old adults. In comparison with similarly derived
data, they also showed that the mean brain weight of men and women aged
20–30 years increased by 109 g (7.5%) and 70 g (5.4%) respectively between
the 1890s and mid twentieth century. A similar but smaller increase was also
seen in the 70–80 year old groups (4% and 3.5%). Another study14 found
an annual increase in brain weight of 0.66 g in men and 0.28 g in women
over the period between 1860 and 1940. These studies demonstrate that the
majority of the difference in brain weight that is attributed to ageing is due to
cohort differences in the populations studied. It is believed that this increase
in brain (and body) size is due to improvements in public health and nutrition
that have occurred during the past century.
Greater than the difference in brain weight between younger and older
adults is the difference between males and females. Men, on average, have a
mean brain weight approximately 100 g greater than women.11,13,14 Differ-
ences in body height accounts for most, but not all, of this difference as brain
weight to body height ratios are not identical for the genders.13 The reasons
for this gender difference are not clear, but it suggests that data from males
and females should be analysed separately in studies of ageing.
Numerous radiological investigations of brain atrophy have been per-
formed. In cross-sectional studies, atrophy has been reported of the whole
brain15,16 as well as specific anatomical regions.17-24 However these findings
have not been supported by longitudinal studies.25-27 In a study of 46 cogni-
tively normal subjects aged 65 and over and followed for between three and
eight years, Mueller et al.26 found minimal increase in ventricular and CSF
volumes and decrease in hippocampus, but no change in other brain regions
examined. The rates of decline were similar for the “young old” (mean age at
commencement of study 70 years) compared to “middle old” (mean age 81
years) and “oldest old” (mean age 87 years) and do not support the concept
of accelerated atrophy in old age. Similarly, Akiyama et al.27 found a small
38 THE AGEING BRAIN

annual increase in the volume of the ventricles and subarachnoid space after
the age of 60 years.
Few quantitative pathological studies of brain volume in normal subjects
have been performed. Miller et al28 reported a decline in both cortical and
white matter volumes after the age of 50 years. After correction for the secu-
lar increase in brain size, the authors found a decline of approximately 2%
per decade. Similarly, a small but significant loss of white matter (approxi-
mately 2 mL/y or 1.3 – 2.2% of cerebrum volume per decade) but no loss of
cerebral cortex volume was demonstrated in subjects aged 46 to 92 years.29
While the analysis of the cortex in its entirety or by lobe, rather than discrete
anatomical regions, may mask subtle decreases in volume, the overall conclu-
sion from these studies is that marked atrophy does not accompany advancing
age.
A variety of factors, other than age, have been shown to accelerate brain
atrophy. Chronic alcohol abuse results in a loss of cerebral white matter30,31
that is partially reversible on abstinence from alcohol. Both cerebral atro-
phy and perfusion have been shown to be influenced by transient ischaemic
attacks, hyperlipidaemia, hypertension, smoking, excessive alcohol consump-
tion and male gender,27 suggesting that much of what is interpreted as age-
related atrophy is indeed due to co-existing risk factors.

Neuronal Loss

As with brain shrinkage, the loss of neurons during ageing is a controversial

issue. Neurons do not regenerate following injury and the exact number of
neurons each person has is not known. Thus is it not possible at death to
accurately determine the degree of neuron loss from any individual. Rather
it is inferred from comparisons with other populations (e.g. younger subjects
or those without disease) or by mathematical extrapolation. Furthermore,
neuronal loss can result from a number of conditions including hypoxia,32
alcohol abuse33,34 and trauma35 and it is essential to exclude such subjects
from studies of ageing.
Several older studies described a marked loss of neurons from the elderly
brain36–39 which was as much as 40% in specific functional regions such
as the frontal cortex.37 In recent years, the methodology employed in these
studies has been criticised. In many instances, neuron counts were performed
using tissue samples which had been processed in paraffin or celloidin, which
causes marked shrinkage. In addition, neuron density was measured, which
does not account for any volume loss of the structure. Newer, unbiased
quantitative techniques have been developed which estimate total neuron
number40,41 and therefore allow more accurate determination of neuronal
loss. To date only one study that rigorously applied these procedures in the
cerebral cortex of normal subjects over a broad spectrum of ages has been
published.42 Interestingly, as with brain size, a difference in neuron number
STRUCTURAL CHANGES 39

with gender which is larger (16%) than any age-effect (10%), and not as a
result of differences in body height, was observed42. The authors quantified
94 brains that were screened to exclude pathological abnormalities, and
found a decrease in total neuron numbers over the range of 20 and 90 years.
However, as this study was performed on the entire cerebral cortex, informa-
tion concerning age effects on specific functional regions of the cortex could
not be derived. Other studies have examined discrete anatomical regions in
a restricted spectrum of ages. Both the superior temporal43 and entorhinal4
cortices appear to have preserved neuronal content between the sixth and
ninth decades. The latter region is of particular interest as it undergoes early
and profound neurodegeneration in AD4 and, along with other parts of the
limbic system, might be expected to show some decline.
Several studies have examined the hippocampal formation in ageing,
although some of these show conflicting results. Neuronal loss from the
subiculum,44,45 CA445 and CA146,47 subregions has been described, although
other studies have shown no loss of CA1 neurons with age.44,45,48 Harding
et al.48 performed a multiple regression analysis and showed that 69% of the
variance in CA1 number is due to brain size while only 2% is due to age. This
study emphasises the difficulties with cross-sectional studies and reinforces
the need to examine other variables in addition to age.
An age-related decline in dendrites and synapses may well be more inform-
ative functionally than a loss of neurons. Structural remodelling (plasticity)
of these is believed to occur throughout life, although it has been shown that
this capacity declines in the oldest old.49 In addition, failure of this adaptive
response has been suggested to underlie neurodegenerative diseases.5 Func-
tional imaging of the ageing brain has yielded conflicting results. Both glo-
bal50 and localised51 decline in cerebral perfusion with age has been reported
in some studies, while others52 find no decline with age once differences in
brain size are corrected for.

β-Amyloid (Aβ) Accumulation

The accumulation of Aβ plaques is one of the hallmarks of AD. However

the high prevalence and density of plaques in many non-demented elderly
subjects is one of the principal factors used to support the ageing and disease
continuum hypothesis. In the cross-sectional autopsy study of Braak and
Braak,53 the proportion of subjects with Aβ plaque deposition ranged from
less than 3% in those 36–40 years to over 75% in those older than 85 years.
These frequencies are based on the examination of a large collection of brains
(N=2,661) from which few exclusions were made (e.g. rare dementing disor-
ders such as progressive supranuclear palsy and corticobasal degeneration).
In addition to the increase in frequency of cases showing plaque deposition,
the proportion which shows the widespread distribution and high density of
plaques usually found in AD (i.e. stage C54) also increases with age. Only
40 THE AGEING BRAIN

three cases below 56 years (0.5%) had stage C plaques, while 30% of cases
above 85 years reached this stage.53 However, epidemiological studies have
found that only around 20% of people over 85 years have dementia55 sug-
gesting that the presence of Aβ plaques is not synonymous with AD.
This suggestion is supported by a number of carefully conducted, pro-
spective studies examining the mismatch between AD-type pathology and
cognitive impairment. Davis et al.56 found that of 59 elderly subjects (mean
age 84 years) without cognitive impairment between 12% and 50% met
neuropathological criteria for AD depending on which criterion was used.
Other cross-sectional57–59 and longitudinal60,61 studies have shown similar
findings. Overall it appears that while Aβ plaque formation is common, it is
not inevitable in the ageing brain and, when present, is not necessarily associ-
ated with clinical dementia.
Part of the controversy associated with the significance of Aβ plaques dur-
ing ageing may stem from the observation that there are several subtypes of
plaque deposits. Four subtypes of plaques have been described.62 Originally
it was thought that these represented stages in the progression of plaque
formation, but more recent studies have suggested that diffuse plaques may
not necessarily progress into neuritic plaques and that neuritic plaques may
develop without evidence of diffuse plaque pathology.63 In a study of 402
subjects aged 30 to 92 years and free from dementia or other neurological
disease, Mackenzie64 showed that while the proportion of subjects with dif-
fuse plaques increased steeply with age, neither the mean nor the maximum
density of plaques increased. This finding suggests that diffuse plaques do not
progressively accumulate in normal ageing, rather that once these plaques
have developed their number remains constant. An increased proportion
of neuritic plaques was found to correlate with age.64 As it is believed that
neuritic plaques rather than diffuse plaques play a role in the pathogenesis of
AD, neuropathological protocols should be modified to incorporate this to
allow the differentiation of normal from pathological ageing.

Neurofibrillary Tangle Accumulation

Neurofibrillary tangles (NFTs) are intracellular accumulations of the micro-

tubule-associated protein tau. Tau is hyperphosphorylated and forms fibrils
in the perikaryon, and later dendrites, of susceptible neurons.65 NFTs are
ultimately lethal and as they are highly insoluble, remain in the neuropil as
extracellular or ghost NFTs.66
The distribution of NFTs in the brain in AD is markedly different to
that of Aβ plaques. NFTs form first in the transentorhinal region, then the
entorhinal, hippocampus and finally the neocortex. This topographic spread
of NFTs is the basis of the Braak staging system for AD pathology.54 In most
cases, the clinical manifestation of dementia corresponds to NFT formation
in the neocortex (stages V and VI). Like Aβ plaques, NFTs are also found
STRUCTURAL CHANGES 41

in the brains of non-demented subjects and their prevalence is higher than

that of Aβ plaques. In the youngest age group studied by Braak and Braak53
(26–30 years), 82% of subjects had NFTs. These were stages I or II, which
correspond to infrequent NFTs in the transentorhinal and entorhinal corti-
ces. By 91–95 years, all subjects have some NFT formation with 24% having
grade V or VI (all of whom also had Aβ plaques). This proportion correlates
well with the prevalence of dementia in this age group,55 which is one of the
reasons why NFTs are considered a better indicator of AD than Aβ plaques.
Other studies support these prevalence figures for NFTs.56, 67
The length of time taken for a NFT to form and to kill a neuron remains
controversial. In one study, it was calculated that it took between three
and five years for a mature NFT to become a ghost NFT, 66 but in another
study it was found that neurons in the CA1 region of the hippocampus
could survive between 15 and 25 years with NFTs. 68 These studies suggest
that the onset of neurofibrillary degeneration may be many years before the
onset of AD, and highlight the difficulties in differentiating normal from
pathological ageing.
The degree of neuron loss due to NFTs has been determined by quantifying
extracellular NFTs. As extracellular NFTs are not readily degraded,66,69,70
they are a good marker of neurodegeneration due to NFTs. In AD, neuro-
nal loss has been found to exceed NFT formation in the temporal43 and
occipital71 cortices, and hippocampus.72 In contrast, NFT formation has been
found to account for all neuronal loss from the nucleus basalis in AD,73 and
from the hippocampus in Parkinson-dementia complex of Guam. 74,75 Similar
studies are yet to be performed in ageing.

Clinicopathological Studies

The relationship between brain pathology and functional impairment is one

of the greatest challenges facing the study of brain ageing. Studies published
to date present conflicting points-of-view with regard to the role of pathology
(especially AD-type pathology) in the causation of functional deficits. There
are many variables which are likely to have an impact on the relationship.
These include factors such as premorbid capacity (brain reserve), genetic sus-
ceptibility and co-existing disease, as well as practical issues such as the range
and type of functional tests performed, the anatomical regions examined, and
the type of pathology identified.
In a study of 59 non-demented subjects, individuals with greater pathology
(“AD-like” using NIA-RI criteria for AD76) had lower scores on specific tests
of immediate and delayed recall than subjects without pathology.77 In addi-
tion, a strong negative correlation has been found between NFT number and
mini-mental state examination (MMSE78) score.79 These, and other similar
studies,80 suggest that the burden of AD-type pathology is closely linked with
cognitive decline.
42 THE AGEING BRAIN

In contrast, a prospective study of 101 subjects aged 75 years and over

revealed a high degree of overlap in the frequency and severity of AD-type
pathology in both demented and non-demented individuals.81 In addition,
no difference on a composite score of neuropsychological tests was found
between individual with and without AD-type pathology.82 Overall these
studies highlight the controversy which still exists concerning the role of AD-
type pathology in age-associated cognitive decline.
Interestingly, the presence of other neurodegenerative pathologies also
increases with age. Davis et al.56 found that only 17% of their subjects were
free from any pathology. Similarly, Xuereb et al.81 showed pathology in a
high proportion of their non-demented subjects including microinfarcts in one
third of cases.

Conclusions

There is little evidence to suggest that major neurodegeneration occurs in

relation to ageing. While minor shrinkage of the white matter has been dem-
onstrated, generalised cortical atrophy has not been identified. In addition,
widespread neuronal loss does not appear to be a feature of normal ageing,
although to date only a small number of functionally-discrete brain regions
have been examined using unbiased methodologies. Cohort differences in
brain size and neuron number have contributed to a misinterpretation of
group comparisons and an over-estimation of age-related decline.
Neurofibrillary tangles and Aβ plaques are central to the pathological
diagnosis of Alzheimer’s disease (AD), but they are also prevalent in the
brains of non-demented elderly subjects. The functional consequences of this
age-related accumulation of AD-type pathology is unclear. Controversies
exist as to whether the presence of plaques and tangles in the brain represents
the earliest manifestations of AD, and whether they are associated with a loss
of neurons and of brain function.
There is a clear need for well designed longitudinal studies of brain ageing
to clarify the relationship between advancing age and neurodegeneration.

Acknowledgements

The author receives funding from the Medical Foundation of The University
of Sydney and the National Health and Medical Research Council of Aus-
tralia. She is a Medical Foundation Fellow.
STRUCTURAL CHANGES 43

References

1. Rubin EH, Storandt M, Miller JP, Kinscherf DA, Grant EA, Morris JC, Berg L.
A prospective study of cognitive function and onset of dementia in cognitively
healthy elders. Arch Neurol. 1998; 55:395–401.
2. Small BJ, Fratiglioni L, Viitanen M, Winblad B, Backman L. The course of cogni-
tive impairment in preclinical Alzheimer disease: three- and 6-year follow-up of
a population-based sample. Arch Neurol. 2000; 57:839–844.
3. Morris JC, McKeel Jr. DW, Storandt M, Rubin EH, Price JL, Grant EA, Ball MJ,
Berg L. Very mild Alzheimer’s disease: informant-based clinical, psychometric
and pathologic distinction from normal aging. Neurology. 1991; 41:469–478.
4. Gomez-Isla T, Price JL, McKeel Jr DW, Morris JC, Growdon JH, Hyman BT.
Profound loss of layer II entorhinal cortex neurons occurs in very mild Alzheim-
er’s Disease. J Neurosci. 1996; 16:4491–4500.
5. Mrak RE, Griffin WST, Graham DI. Aging-associated changes in human brain.
J Neuropathol Exp Neurol. 1997; 56:1269–1275.
6. Perls TT. The oldest old. Sci Am. 1995; 272:50–55.
7. Di Patre PL, Read SL, Cummings JL, Tomiyasa U, Vartavarian LM, Secor DL,
Vinters HL. Progression of clinical deterioration and pathological changes in
patients with Alzheimer disease evaluated at biopsy and autopsy. Arch Neurol.
1999; 56:1254–1261.
8. Bennett DA, Cochran EJ, Saper CB, Leverenz JB, Gilley DW, Wilson RS. Patho-
logical changes in frontal cortex from biopsy to autopsy in Alzheimer’s disease.
Neurobiol Aging. 1993; 14:589–596.
9. Mann DMA, Marcyniuk B, Yates PO, Neary D, Snowden JS. The progression of
the pathological changes of Alzheimer’s disease in frontal and temporal neocortex
examined both at biopsy and autopsy. Neuropathol Appl Neurobiol. 1988; 14:
177–195.
10. Martin EM, Wilson RS, Penn RD, Fox JH, Clasen RA, Savoy SM. Cortical biopsy
results in Alzheimer’s disease: correlation with cognitive deficits. Neurology.
1987; 37:1201–1204.
11. Samorajski T. How the human brain responds to aging. J Am Geriatr Soc. 1976;
24:4–11.
12. Ho K, Roessmann U, Straunfjord JV, Monroe G. Analysis of brain weight. I.
Adult brain weight in relation to sex, race, and age. Arch Path Lab Med. 1980;
104:635–639.
13. Dekaban AS, Sadowsky D. Changes in brain weights during the span of human
life: relation of brain weights to body heights and body weights. Ann Neurol.
1978; 4:345–356.
14. Miller AKH, Corsellis JAN. Evidence for a secular increase in human brain weight
during the past century. Ann Human Biol. 1977; 4:253–257.
15. Yue NC, Arnold AM, Longstreth Jr. WT, Elster AD, Jungreis CA, O’Leary DH,
Poirier VC, Bryan RN. Sulcal, Ventricular and white matter changes at MR imag-
ing in the aging brain: data from the cardiovascular health study. Radiology.
1997; 202:33–39.
16. Murphy DG, de Carli C, Schapiro MB, Rapoport SI, Horwitz B. Age-related dif-
ferences in volumes of subcortical nuclei, brain matter and cerebrospinal fluid in
heathly men as measured with magnetic resonance imaging. Arch Neurol. 1992;
49:839–845.
17. Coffey CE, Wilkinson WE, Parashos IA, Soady SAR, Sullivan RJ, Patetrson LJ,
Figiel GS, Webb MC, Spritzer CE, Djang WT. Quantitative cerebral anatomy of
the aging human brain: A cross-sectional study using magnetic resonance imag-
ing. Neurology. 1992; 42:527–536.
44 THE AGEING BRAIN

18. Coffey CE. Anatomic imaging of the aging human brain: computered tomography
and magnetic resonance imaging. In: Coffey CE, Cummings JL, editors. Textbook
of Geriatric Neuropsychiatry. Washington, DC: American Psychiatric Press Inc,
1994; 159–194.
19. Coffey CE, Lucke JF, Saxton JA, Ratcliff G, Unitas LJ, Billig B, Bryan RN. Sex
differences in brain aging: a quantitative magnetic resonance imaging study. Arch
Neurol. 1998; 55:169–179.
20. Convit A, de Leon MJ, Tarshish C, de Santi S, Kluger A, Rusinek H, George AE.
Hippocampal volume losses in minimally impaired elderly. Lancet. 1995; 345:
266.
21. Convit A, de Leon M, Hoptman MJ, Tarshish C, De Santi S, Rusinek H. Age-
related changes in brain: I. Magnetic resonance imaging measures of temporal
lobe volumes in normal subjects. Psychiat Quart. 1995; 66:343–455.
22. Raz N, Torres IJ, Spencer WD, White K, Acker JD. Age-related regional differ-
ences in cerebellar vermis observed in vivo. Arch Neurol. 1992; 49:412–416.
23. Raz N, Gunning FM, Head D, Dupuis JH, McQuain J, Briggs SD, Loken WJ,
Thornton AE, Acker JD. Selective aging of the human cerebral cortex observed in
vivo: differential vulnerability of the prefrontal gray matter. Cereb Cortex. 1997;
7:268–282.
24. Raz N, Dupuis JH, Briggs SD, McGavran C, Acker JD. Differential effects of age
and sex on the cerebellar hemispheres and the vermis: a prospective MR study.
AJNR. 1998; 19:65–71.
25. Fox NC, Cousens S, Scahill R, Harvey RJ, Rossor MN. Using serial registered
brain magnetic resonance imaging to measure disease progression in Alzheimer
disease: power calculations and estimates of sample size to detect treatment
effects. Arch Neurol. 2000; 57:339–344.
26. Mueller EA, Moore MM, Kerr DC, Sexton G, Camicioli RM, Howieson DB,
Quinn JF, Kaye JA. Brain volume preserved in healthy elderly through the elev-
enth decade. Neurology. 1998; 51:1555–1562.
27. Akiyama H, Meyer JS, Mortel KF, Terayama Y, Thornby JI, Konno S. Normal
human aging: factors contributing to cerebral atrophy. J Neurol Sci. 1997; 152:
39–49.
28. Miller AKH, Alston RL, Corsellis JAN. Variation with age in the volumes of
grey and white matter in the cerebral hemispheres of man: measurements with
an image analyser. Neuropathol Appl Neurobiol. 1980; 6:119–132.
29. Double KL, Halliday GM, Kril JJ, Harasty JA, Cullen K, Brooks WS, Creasey
H, Broe GA. Topography of brain atrophy during normal aging and Alzheimer’s
disease. Neurobiol Aging. 1996; 17:513–521.
30. Harper CG, Kril JJ, Holloway RL. Brain shrinkage in chronic alcoholics — a
pathological study. Br Med J. 1985; 290:501–504.
31. Kril JJ, Halliday GM. Brain shrinkage in alcoholics: a decade on and what have
we learned? Prog Neurobiol. 1999; 58:381–387.
32. Auer RN, Benveniste H. Hypoxia and related conditions. In: Graham DI, Lantos
PL, editors. Greenfield’s neuropathology. London: Arnold, 1997; 263–314.
33. Kril JJ. The contribution of alcohol, thiamine deficiency and cirrhosis of the liver
to cerebral cortical damage in alcoholics. Metab Brain Dis. 1995; 10:9–16.
34. Kril JJ, Halliday GM, Svoboda MD, Cartwright H. The cerebral cortex is dam-
aged in chronic alcoholics. Neuroscience. 1997; 79:983–998.
35. Graham DI, Gennarelli TA. Trauma. In: Graham DI, Lantos PL, editors. Green-
field’s neuropathology. London: Arnold, 1997; 197–262.
36. Anderson JM, Hubbard BM, Coghill GR, Slidders W. The effect of advancing age
on the neurone content of the cerebral cortex. J Neurol Sci. 1983; 58:233–244.
37. Brody H. Organisation of the cerebral cortex III. A study of aging in the human
cerebral cortex. J Comp Neurol. 1955; 102:511–556.
STRUCTURAL CHANGES 45

38. Henderson G, Tomlinson BE, Gibson PH. Cell counts in human cerebral cortex
in normal adults throughout life using an image analysing computer. J Neurol
Sci. 1980; 46:113–136.
39. Terry RD, de Theresa R, Hansen LA. Neocortical cell counts in normal human
adult aging. Ann Neurol. 1987; 21:530–539.
40. Gundersen HJG, Bendtsen TF, Korbo L, Marcussen N, Moller A, Nielsen K,
Nyengaard JR, Pakkenberg B, Sorensen FB, Vesterby A, West MJ. Some new,
simple and efficient stereological methods and their use in pathological research
and diagnosis. APMIS. 1988; 96:379–394.
41. Gundersen HJG, Bagger P, Bendtsen TF, Evans SM, Korbo L, Marcussen N,
Moller A, Nielsen K, Nyengaard JR, Pakkenberg B, Sorensen FB, Westerby A,
West MJ. The new stereological tools: disector, factionator, nucleator and point
sampled intercepts and their use in pathological research and diagnosis. APMIS.
1988; 96:857–881.
42. Pakkenberg B, Gundersen HJG. Neocortical neuron number in humans: effects
of sex and age. J Comp Neurol. 1997; 384:312–320.
43. Gomez-Isla T, Hollister R, West H, Mui S, Growdon JH, Petersen RC, Parisi JE,
Hyman BT. Neuronal loss correlates with but exceeds neurofibrillary tangles in
Alzheimer’s disease. Ann Neurol. 1997; 41:17–24.
44. West MJ. Regionally specific loss of neurons in the aging human hippocampus.
Neurobiol Aging. 1993; 14:287–293.
45. West MJ, Coleman PD, Flood DG, Troncoso JC. Differences in the pattern of
hippocampal neuronal loss in normal ageing and Alzheimer’s disease. Lancet.
1994; 344:769–772.
46. West MJ, Gundersen HJG. Unbiased stereological estimation of the number of
neurons in the human hippocampus. J Comp Neurol. 1990; 296:1–22.
47. Simic G, Kostovic I, Winblad B, Bogdanovic N. Volume and number of neurons
of the human hippocampal formation in normal aging and Alzheimer’s disease. J
Comp Neurol. 1997; 379:482–494.
48. Harding AJ, Halliday GM, Kril JJ. Variation in hippocampal neuron number with
age and brain volume. Cereb Cortex. 1998; 8:710–718.
49. Flood DG, Buell SJ, Defiore CH, Horwitz GJ, Coleman PD. Age-related dendritic
growth in dentate gyrus of human brain is followed by regression in the “oldest
old”. Brain Res. 1985; 345:366–368.
50. Petit-Taboue MC, Landeau B, Desson JF, Desgranges B, Baron JC. Effects of
healthy aging on the regional cerebral metabolic rate of glucose assessed with
statistical parametric mapping. Neuroimage. 1998; 7:176–184.
51. Schultz SK, O’Leary DS, Boles Ponto LL, Watkins GL, Hichwa RD, Andreasen
NC. Age-related changes in regional cerebral blood flow among young to mid-life
adults. Neuroreport. 1999; 10:2493–2496.
52. Meltzer CC, Cantwell MN, Greer PJ, Ben-Eliezer D, Smith G, Franj G, Kaye WH,
Houck PR, Price JC. Does cerebral blood flow decline in healthy aging? A PET
study with partial-volume correction. J Nucl Med. 2000; 41:1849–1848.
53. Braak H, Braak E. Frequency of stages of Alzheimer-related lesions in different
age categories. Neurobiol Aging. 1997; 18:351–357.
54. Braak H, Braak E. Neuropathological staging of Alzheimer-related changes. Acta
Neuropathol. 1991; 82:239–259.
55. Jorm AF. The epidemiology of Alzheimer’s disease and related disorders. London:
Chapman and Hall, 1990.
56. Davis DG, Schmitt FA, Wekstein DR, Markesbery WR. Alzheimer neuropatho-
logic alterations in aged cognitively normal subjects. J Neuropathol Exp Neurol.
1999; 58:376–388.
57. Price JL, Davis PB, Morris JC, White DL. The distribution of tangles, plaques and
related immunohistochemical markers in healthy aging and Alzheimer’s disease.
Neurobiol Aging. 1991; 12:295–312.
46 THE AGEING BRAIN

58. McKee AC, Kosik KS, Kowall NW. Neuritic pathology and dementia in Alzhe-
imer’s disease. Ann Neurol. 1991; 30:156–165.
59. Giannakopoulos P, Hof PR, Mottier S, Michel JP, Bouras C. Neuropathological
changes in the cerebral cortex of 1258 cases from a geriatric hospital: retrospec-
tive clinicopathological evaluation of a ten year autopsy population. Acta Neu-
ropathol. 1994; 87:456–468.
60. Dickson DW, Crystal HA, Mattiace LA, Masur DM, Blau AD, Davies P, Yen S-H,
Aronson MK. Identification of normal and pathological aging in prospectively
studied nondemented elderly humans. Neurobiol Aging. 1991; 13:179–189.
61. Price JL, Morris JC. Tangles and plaques in nondemented aging and “Preclinical”
Alzheimer’s disease. Ann Neurol. 1999; 45:358–368.
62. Delaere P, Duyckaerts C, Piette F, Hauw JJ. Subtypes and differential laminar
distributions of BA4 deposits in Alzheimer’s disease: Relationship with the intel-
lectual status of 26 cases. Acta Neuropathol. 1991; 81:328–335.
63. Armstrong RA. Relationships between morphological types of plaque in Alzhe-
imer’s disease as revealed in silver and immunostained preparations. Neurode-
generation. 1993; 2:259–266.
64. Mackenzie IRA. Senile plaques do not progressively accumulate with normal
aging. Acta Neuropathol. 1994; 87:520–525.
65. Bancher C, Brunner C, Lassmann H, Budka H, Jellinger K, Wiche G, Seitelberger
F, Grundke-Iqbal I, Iqbal K, Wisniewski HM. Accumulation of abnormally phos-
phorylated tau precedes the formation of neurofibrillary tangles in Alzheimer’s
disease. Brain Res. 1989; 477:90–99.
66. Bobinski M, Wegiel J, Tarnawski M, Bobinski M, De Leon MJ, Reisberg B, Miller
DC, Wisniewski HM. Duration of neurofibrillary changes in the hippocampal
pyramidal neurons. Brain Res. 1998; 799:156–158.
67. Duyckaerts C, Hauw JJ. Prevalence, incidence and duration of Braak’s stages in
the general population: can we know? Neurobiol Aging. 1997; 18:362–369.
68. Morsch R, Simon W, Coleman PD. Neurons may live for decades with neurofi-
brillary tangles. J Neurol Neurosur Ps. 1999; 58:188–197.
69. Cras P, Smith MA, Richey PL, Siedlak SL, Mulvihill P, Perry G. Extracellular
neurofibrillary tangles reflect neuronal loss and provide further evidence of
extensive protein cross-linking in Alzheimer disease. Acta Neuropathol. 1995;
89:291–295.
70. Braak H, Braak E. Evolution of neuronal changes in the course of Alzheimer’s
disease. J Neural Transm Suppl. 1998; 53:127–140.
71. Leuba G, Kraftsik R. Visual cortex in Alzheimer’s disease: occurrence of neuronal
death and glial proliferation, and correlation with pathological hallmarks. Neu-
robiol Aging. 1994; 15:29–43.
72. Kril JJ, Patel S, Harding AJ, Halliday GM. Neuron loss from the hippocampus of
Alzheimer’s disease exceeds extra cellular neurofibrillary tangle formation. Acta
Neuropathol. 2002; 103:370–376.
73. Cullen KM, Halliday GM. Neurofibrillary degeneration and cell loss in the
nucleus basalis in comparison to cortical Alzheimer pathology. Neurobiol Aging.
1998; 19:297–306.
74. Schwab C, Schulzer M, Steele JC, McGeer PL. On the survival time of a tangled
neuron in the hippocampal CA4 region in parkinsonian dementia complex of
Guam. Neurobiol Aging. 1999; 20:57–63.
75. Schwab C, Steele JC, McGeer PL. Pyramidal neuron loss is matched by ghost tan-
gle increase in Guam parkinsonism-dementia hippocampus. Acta Neuropathol.
1998; 96:409–416.
76. National Institute on Aging and Reagan Institute Working Group on diagnostic
criteria for the neuropathological diagnosis of Alzheimer’s disease. Consensus
recommendations for the postmortem diagnosis of Alzheimer’s disease. Neurobiol
Aging. 1997; 18(4S):S1–S3.
STRUCTURAL CHANGES 47

77. Schmitt FA, Davis DG, Wekstein DR, Smith CD, Ashford JW, Markesbury WR.
“Preclinical” AD revisited: Neuropathology of cognitively normal older adults.
Neurology. 2000; 55:370–376.
78. Folstein MF, Folstein SE, McHugh PR. “Mini-mental”: A practical method for
upgrading the cognitive state of patients for the clinician. J Psychiat Res. 1975;
12:189–198.
79. Green MS, Kaye JA, Ball MJ. The Oregan brain aging study. Neuropathology
accompanying healthy aging in the oldest old. Neurology, 2000; 54:105–113.
80. Morris JC, Storandt M, McKeel Jr. DW, Rubin EH, Price JL, Grant EA, Berg L.
Cerebral amyloid deposition and diffuse plaques in “normal” aging: evidence
for presymptomatic and very mild Alzheimer’s disease. Neurology. 1996; 46:
707–719.
81. Xuereb JH, Brayne C, Dufouil C, Gertz H, Wischik C, Harrington C, Mukaetova-
Ladinska E, McGee MA, O’Sullivan A, O’Connor D, Paykel ES, Huppert FA.
Neuropathological findings in the very old. Results from the first 101 brains of a
population-based longitudinal study of dementing disorders. Ann NY Acad Sci.
2000; 903:490–496.
82. Crystal HA, Dickson D, Sliwinski M, Masur D, Blau A, Lipton RB. Associations
of status and change measures of neuropsychological function with pathologic
changes in elderly, originally nondemented subjects. Arch Neurol. 1996; 53:
82–87.
Chapter 4

STRUCTURAL
NEUROIMAGING OF
THE AGEING BRAIN
Jeffrey C.L. Looi and Perminder S. Sachdev*

Introduction

The brain undergoes progressive changes in adult life, with these changes
accelerating in the later years. The availability of neuroimaging techniques
has opened up the possibility of studying these changes in large representa-
tive samples of elderly individuals, something not possible with post-mortem
examinations. The introduction of computerised axial tomography (CT) was
a major advance in the 1970s, but magnetic resonance imaging (MRI) has
enhanced this capacity immeasurably. Structural neuroimaging provides data
for structural-functional correlation in both healthy and diseased individuals,
and enables a comparison of normative age-related changes with those due
to neurodegenerative and other disorders. It also helps us understand which
aspects of brain changes in the elderly are truly related to ageing, and which
may in fact be due to age-related diseases. A differentiation of these is crucial
for developing strategies for intervention.
Advances in imaging technology have been paralleled by significant devel-
opments in the techniques of qualitative and quantitative analysis of neuroim-
aging. Similarly, research has progressed from cross-sectional to longitudinal
studies, using repeated imaging to characterise the evolving changes in indi-
vidual subjects. Such research has limitations due to differential selection
criteria, sample sizes, imaging modality utilised, measurement methods and
duration of longitudinal follow-up. Nonetheless, there are some consistent

*To whom correspondence should be addressed.

50 THE AGEING BRAIN

findings in gross measures of brain volume/atrophy as well as for individual

regions of the brain which will be reviewed in this chapter. These changes
concord, in many cases, with known neuropathological and neuropsychologi-
cal data on the ageing brain.

Limitations of the Studies

Before we summarise the studies, we would like to outline some limitations

of the data currently available. The structural imaging literature spans many
different iterations in technology and technique. The focus of our review has
been on the more recent rigorously designed CT and MRI studies. The dif-
ferentiation between normal (i.e., with no overt neurological symptoms) and
successful ageing (i.e., with minimal physiologic loss even when compared
with younger individuals) is a difficult task. 1,2 Due to survivor effects, studies
may contain an over-representation of very healthy individuals.1,2 Most series
of neuroimaging studies of the ageing brain describe findings from individu-
als without overt neurological dysfunction, but do not necessarily analyse
all risk factors for impairment or have indices of normal function such as
neuropsychological tests or neurological function.1 The degree of ascertain-
ment of neurological status has been extremely variable and not generally
satisfactory, varying from use of limited screening to full neurological exami-
nation. These assessments have usually not included detailed neuropsycho-
logical assessment. When these have been performed, they have depended on
patient groups or healthy volunteers, rather than a randomly selected group
which might reflect a community-dwelling population.3 Some studies utilise
qualitative rating scales of atrophy, white matter change, etc., whilst others
utilise semi-automated quantitative methods based upon delineation of the
area of interest and calculation of the volume or degree of change occurring.
The types of data obtained from these methods are necessarily different.
The sample size of individual studies may limit the ability to discern dif-
ferences across the age range, as the majority of studies comprise groups of
50–100 individuals across the entire age range from youth to old age. There-
fore, the subgroup of the aged may contain only 10–25 subjects, reducing
the power of the study to demonstrate age effects. We have therefore concen-
trated on reviewing larger studies and those assessing longitudinal change in
aged cohorts. Furthermore, cross-sectional studies may be subject to cohort
effects from changes in nutrition, living conditions and health care when older
persons are compared with contemporary youth.2 Longitudinal studies may
be affected by obsolescence of imaging technology.2
Finally, MRI yields better differentiation between grey and white matter,
does not expose the subject to radiation (obviating the concern about serial
radiation exposure), allows for facility of image acquisition and manipulation
within various anatomical planes, and may be more able to detect pathologi-
cal change.2
STRUCTURAL NEUROIMAGING 51

The Neuroimaging Studies of Normal Ageing

Atrophy and ventricular enlargement

Computerised tomography studies

A large number of studies have used CT to determine changes in brain volume
and indices of brain atrophy with age. Using an automated method to count
pixels on the CT scan, the ratio of brain to the cranial cavity volume ( the
cranio-cerebral index, CCI) was calculated for 228 healthy volunteers.4 The
CCI for those aged 20–49 was used as the baseline for comparison as an indi-
cator of brain atrophy, yielding a brain volume index (BVI) which decreased
markedly after 50 years.4 However, this may be artefactually related to the
cut-off of 49 used for the baseline. The ventricle to brain ratio did not change
significantly for a subset of 93 subjects in a study to age 40, nor were changes
noted in grey/white matter or total brain density calculated in Hounsfield units
(HU).5 Ventricular enlargement was investigated using a semi-automated com-
puter analysis in 123 normal subjects aged 23–88, showing only slight increases
in ventricular volume to age 60, but more rapid increase in size thereafter.6
A large study of 980 healthy volunteers aged 10 to 88 years used CT to
measure CSF and cranial cavity volumes and calculate a brain atrophy index
(BAI) from their ratio.7 Both CSF volume and BAI increased significantly after
thirties in men and women, with the rate of decline in males in the 30–40s
age range being twice that of the women, slowing after this decade. In a large
series of 381 healthy volunteers, the brain atrophy index (BAI) was calculated
as above, finding that both CSF space volume and BAI were smallest in the
30s for men and 20s for women, but that significant increases occurred after
age 40 in both sexes, with more marked atrophy occurring after 40s in men
and 50s in women.8
In a study of 152 asymptomatic subjects aged 17–86 in Japan, the ven-
tricular volume was noted to increase in males from age 40 onwards and in
women from 50 onwards, with CSF space volume increasing from the 40s in
men and 60s in women.9 A large series of 212 normal elderly persons found
that ventricular dilatation increased in a statistically significant manner
with increased age, but that cortical atrophy was non-significant.10 Analys-
ing midventricular, high ventricular and supraventricular slices for percent
fluid volume and mean CT density for each slice, it was concluded that fluid
volume at the ventricular level was stable until the 60s, whilst the high and
supraventricular volumes increased in the 50s in 79 men aged 31–87.11 This
study was notable for careful screening for health, the subjects representing
a subset of a normative ageing study.

Magnetic Resonance Imaging studies

In a study of 76 healthy adults, increasing age was associated with increas-
ing volume of the third ventricle (2.8% per year) and lateral ventricles (3.2%
per year).12 A significant positive correlation of ventricle to brain ratio with
52 THE AGEING BRAIN

increasing age was found in a small sample.13 In a group of 142 healthy vol-
unteers aged 21–80, lateral ventricular volume increased significantly with
age, ranging from 134% in males and 66% in females in comparison between
subgroups 21–30 to 71–80.14 The Helsinki Aging Brain study investigated 128
randomly selected, community-dwelling, neurologically non-diseased elderly
(as confirmed by neurological examination), and found that central (enlarge-
ment of the lateral and third ventricles) atrophy increased significantly with
age after stratifying the cohorts into the young-old (56–72) and the old-old
(77–88).3 Decreasing total brain volume, increasing subarachnoid CSF, third
and lateral ventricle volume have been observed, with the differences being
more pronounced for males, especially in the third ventricle (194 subjects).15
In a study of 69 healthy subjects aged 20–85, there were significant
increases with age in ventricular and peripheral CSF volume.16 A long term
five-year follow-up study of 24 successfully aged elderly without significant
neuropyschometric decline showed a modest increase in volumes of the
lateral ventricles and frontal CSF spaces.17 Ventricular enlargement was
significantly associated with age and sex in a sample of 3660 community
dwelling elderly.18 A sample of 330 healthy elderly was recruited from the
multicentric Cardiovascular Health Study and it was found that increased age
was associated with increased volumes of the lateral and third ventricles.19 In
a study utilising quantitative methods to assess various brain parameters in
116 subjects aged 19 months to 80 years, total CSF brain volume was found
to increase linearly with age, whilst as a percentage of intracranial space
volume it rated 7–9% up to adolescence, increasing from the 2nd decade to
range 20–30% in those 71–80.20 The Baltimore Longitudinal Study of Aging
investigated 116 subjects, finding that ventricular volumes and ventricle to
brain ratios increased significantly with age and sex (males showing more
atrophy).21 Longitudinal analysis of ventricular volume showed a decrease of
1525.6 mm3 in ventricular volume (0.5% of total brain volume) over a year
and change in VBR of 0.0016, yet the authors stated that these measures were
repeatable and showed stability within there brain measurements.21

Conclusion
Studies investigating ventricular enlargement have used qualitative rating
scales, semi-automated or automated measurement of ventricular volume and
the calculation of ventricle to brain ratio. There is considerable heterogeneity
in the findings, yet some firm conclusions can be drawn. Both CT and MRI
studies demonstrate ventricular enlargement with increasing age, indicating
that enlargement begins as early as the 30s in males and the 40s in females,
with males showing earlier and more marked atrophy.

Gross brain volume

Computerised tomography studies
Even in a group of 115 subjects aged up to only 40, increased sulcal widening
of the frontal lobes and cerebellar vermis beginning in teenage years contin-
STRUCTURAL NEUROIMAGING 53

ued to progress.5 Computerised quantitative estimation of volume percentage

of brain to cranial cavity based on Hounsfield values showed that in men,
brain atrophy began in the sixth decade and continued to decline through to
the eighth decade.22 In women, decline in volume began in the fifth decade
and appeared to plateau in fifth to sixth decades, to decline again in the sev-
enth and eighth decades. A large series of 212 normal elderly persons found
that, with increased age, increase in cortical atrophy was non-significant with
a trend towards greater increase in males than females.10
Brain atrophy is usually a diffuse process and attempts have been made to
derive linear parameters that describe the atrophy. This yielded a model with
combined measurements of the lateral and third ventricles, Sylvian fissure and
pre-pontine cistern to describe atrophy.23 A study of 64 healthy males aged
31–87 demonstrated bilateral, symmetrical atrophy of the cingulate gyrus
and sulcus, interhemispheric frontal gyri and parieto-occipital sulcus.24 This
study also showed asymetrical widening of the central and postcentral sulcus
on the left and the intraparietal sulcus on the right.

Magnetic resonance imaging studies

In a study of 76 healthy adults, increasing age was associated with decreasing
volume of cerebral hemispheres (0.23% per year), frontal lobes, (0.55% per
year), temporal lobes (0.28% per year) and amygdala–hippocampal complex
(0.30% per year).12 In a group of 142 healthy volunteers aged 21–80, cer-
ebral hemispheric volume decreased significantly with age.14 In a study of 69
healthy subjects aged 20–85, a subgroup of a PET study, showed significant
decrements in brain matter volumes of cerebral hemispheres, frontal lobe, the
parieto-occipital lobe and parahippocampal gyrus.16 Men had a significantly
greater age-related decrease than women in brain volume in the cerebral
hemispheres, and frontal and temporal lobes in this study. Left frontal regions
decreased more in women and right frontal regions decreased more in men.
A sample of 330 healthy elderly individuals was recruited from the multi-
centric Cardiovascular Health Study and it was found that increased age was
associated with decreased cerebral hemispheric, frontal and temporo-parietal
volumes.19 The Baltimore Longitudinal Study of Aging investigated 116 sub-
jects, finding that gross brain volumes decreased significantly with age and sex
(males showing more atrophy).21 Additionally, younger persons had signifi-
cantly larger grey than white matter volumes.21 An analysis of regional brain
volume in the same study yielded a significant region by hemisphere interac-
tion, reflecting greater right than left frontal, and left compared with right
temporal volumes. Investigation of gross brain volume of various regions (331
subjects) standardised for cranial size showed markedly smaller volumes for
frontal, temporal lobes, basal ganglia, thalamus, and parietal/parieto-occipital
lobes, with sparing of only the basal frontal lobe and left cerebellum with age-
ing.25 In the posterior right frontal lobe, males demonstrated more atrophy
then females. Males demonstrated atrophy in the right temporal, left basal
ganglia, parietal lobe and cerebellum in contrast to females.
54 THE AGEING BRAIN

In a study utilising quantitative methods to assess various brain param-

eters in 116 subjects aged 19 months to 80 years, whole brain volume stead-
ily declined from 16–80 years, such that by 71–80 it had declined by 26%
smaller than healthy 2–3 year olds.20 Whole brain volumes were 12% smaller
in female volunteers.20

Conclusion
Gross brain volumes decrease with age. Whilst there has been considerable
variation in the brain volumes studied, there is consistent evidence of regional
decreases in frontal and temporo-parietal volumes. Again, males show earlier
atrophy and greater rates of decline.

Cortex

Computerised tomography study

Quantitation of cerebral compartment densities was made via measurement
of Hounsfield unit (HU) values in 81 healthy volunteers across an age range,
yielding decline in HU values within all cortical grey matter and fronto-pari-
etal white matter.26 Compartmental volumes for cortical and subcortical grey
matter declined with increasing age.26

MRI studies
In a study of 76 healthy adults, increasing age was associated with increasing
odds of cortical atrophy at the rate of 8.9% per year.12 Assessment of 154
medically and cognitively healthy elderly individuals (55-88) found that 33%
of the subjects had evidence of hippocampal atrophy (58% bilateral) and the
prevalence for the same increased with age (12.8% in 55–65 to 56.8% in
77–88), with changes more common in males.27 In a study of 3660 commu-
nity–living elderly, sulcal atrophy independently and significantly increased
with age and sex (male), and was associated with ventricular atrophy.18 In a
study of 619 healthy volunteers, the volume of the hippocampal formation
and the amygdala reduced gradually with increasing age, with some increase
in the volume of the temporal horn especially in those over 61.28
In a study of 126 healthy control subjects for an AD study, normalised
medial temporal lobe (MTL) volumes were found to decrease with age in a
linear fashion, with the greatest decline being in the head of the hippocam-
pus.29 Women had larger normalised MTL volumes than men. The mean
volumetric decline (cubic mm) was 45.63 for the total hippocampus, 27.43
for the hippocampal head, 8.484 for the hippocampal body, 9.68 for the
hippocampal tail, 46.65 for the parahippocampal gyrus and 20.75 for the
amygdala.29 Mean total intracranial volume was 1,393 ± 133 cubic cm. In
a group of 24 cognitively normal subjects, the same group found the annual
average rate of hippocampal volume loss was 75 ± cubic mm or –1.55
± 1.38% and was greater for the head then the rest of the hippocampus.30 A
low steady rate of hippocampal atrophy (-1.73%) per year over four years
STRUCTURAL NEUROIMAGING 55

was found for control subjects with stable cognition in an investigation of

hippocampal atrophy as a predictor of mild cognitive impairment or AD.31
In a study utilising quantitative methods to assess various brain parameters
in 116 subjects aged 19 months to 80 years, grey matter volume was found
to decrease by 5% per decade throughout life.20
One of the problems in analysing for cortical atrophy has been the highly
gyrified cerebral cortex in humans that presents problems in quantifying
the volume of cortex.32 This is further complicated by the sulcal and gyral
changes that may be maturational as opposed to involutional. Some groups
have made attempts to characterise the complex structure of the cortical sur-
face by using a surface-rendering program that can generate a surface mask
for use as a comparator with study subjects. However, these surface maps
are not able to quantify sulcal invaginations filled with CSF, and investiga-
tors at the University of Iowa have developed a program (BRAINSURF) to
more accurately map the cortical surface and hence calculate volume.32 They
describe a sample of 148 healthy elderly in which they noted gyral, sulcal and
cortical thickness changes with ageing. The cortical mantle becomes thin-
ner with increasing age, more rapidly in males, and sulcal curvature index
becomes larger, reflecting widening of the sulci.32 The comparisons were valid
for younger (<20 years) versus older (>40 years) subjects.

Conclusion
There are technical problems in calculating cortical atrophy related to the
complex gyral/sulcal structure and orientation of various cortical regions. To
an extent, the superior imaging characteristics of MRI have allowed descrip-
tion and quantitation of changes that were undetectable with earlier imaging
technologies. Age-related decreases in hippocampal and parahippocampal
volumes have been well–characterised with MRI, yielding rates of decline
of 5-10% per decade. Whilst new surface-mapping techniques may assist
in detection of atrophy, there remains the problem of measurement error
obscuring age-related changes, given the slow rate of atrophy noted.

Subcortical structures

Deep white matter

Computerised tomography study. In a group of 123 normal subjects aged 23
to 88 years, there was a significant negative correlation between Hounsfield
Score attenuation values and age, and this effect was enhanced after the effect
of cranial size was partialled out.33 Furthermore, no significant age relation-
ship could be found in those under the age of 56 in this sample and no focal
or generalised radiolucency was noted in the older subjects.33

MRI studies. Examining 23 formalin-fixed brain specimens of patients 60

years and older, subtle changes of gliosis and demyelination accounted for
the majority of hyperintense white matter lesions seen in MRI. 34 A signifi-
56 THE AGEING BRAIN

cant increase of lacunar infarction was noted with increasing age in a small
sample, with 34% of the lesions occurring in the deep white matter. 13 In
a group of 142 healthy volunteers aged 21–80, white matter hyperintensi-
ties increased almost linearly with age, ranging from 20% in those 21–30,
to 100% in those 71–80 years old.14 The Helsinki Aging Brain study
investigated 128 randomly selected, community-dwelling, neurologically
non-diseased elderly (as confirmed by neurological examination), stratify-
ing the cohort into the young-old (56–72) and the old-old (77–88). 3 This
study found that periventricular hyperintensities were present in 21% and
65% respectively.3 In the same sample, centrum semiovale hyperintensities
were 11% and 38% respectively. These findings for hyperintensities were
significantly associated with central (third and lateral ventricular) atrophy.
There was a significant, non-linear, increase in periventricular hyperintensity
with age, but not with gender.3 A T2-weighted MRI series of 61 healthy
volunteers aged 30-86 showed that age-related changes emerged after 50,
comprising increased signal intensity in white matter, high signal foci and
constant high signal foci after the age of 65, whilst the grey matter remained
stable.35
The Cardiovascular Health Study investigated white matter changes in
3301 elderly people (65 or older), finding that only 4.4% were free of any
findings.36 The majority (80%) of these changes were classed as mild and
comprised mainly symmetrical, supratentorial or periventricular changes,
but higher grades were significantly associated with greater age. Longstreth
et al.36 also noted one third of the group had suffered silent strokes. Another
analysis of the same cohort showed the mean white matter grade increased
with age and was associated with ventricular grade, and found that 34.7%
had little or no white matter changes.18 The predominant distribution of
white matter change was periventricular in 72.7%. 18 In a study utilising
quantitative methods to assess various brain parameters in 116 subjects
aged 19 months to 80 years, white matter volume reached a plateau in the
4th decade with reduction of 13% in those aged 70–80 years. 20 The recent
Rotterdam MRI study of 1077 subjects aged 60–90 years showed 8% of all
subjects had no subcortical white matter lesion, 20% had no periventricular
lesion, and 5% had no white matter lesions in either location. 37 Thus, the
prevalence of white matter lesions in the aged was high and the proportion
increased with age, with trends toward more subcortical lesions in men in
frontal white matter and periventricular lesions in women. 37
A five year follow-up study of 24 successfully aged elderly without sig-
nificant neuropyschometric decline showed a modest increase in deep white
matter hyperintensities with increasing age.17 Similarly, lesion progression
was found in 17.9% of 273 healthy elderly volunteers in the Austrian stroke
prevention study, with 64.5% of the volunteers (mean age 60+/–6.1 years)
having white matter hyperintensities at baseline. 38 At baseline, punctate
WMH occurred in 52%, and confluent changes in 12.5%.
STRUCTURAL NEUROIMAGING 57

Conclusion
White matter hyperintensities in the periventricular regions are predominant
in the elderly, with rates approaching 100% in large longitudinal studies,
though the majority of changes are mild. Lesion progression has also been
demonstrated. Quantitation of the volume of WMHs is needed to assist in
characterising the rate of change and differentiating age-related changes from
neurodegenerative disease. Subcortical changes are less common but have a
prevalence of approximately 20%.

Corpus callosum and midline structures

MRI studies
In a small sub-sample of 56 persons aged 16–60 years, a decrease in size of
the pituitary gland and corpus callosum was noticeable in the 51–60 year
old group, whilst no such declines were evident in the pons or the cerebellar
vermis39. Age negatively correlated with corpus callosum cross-sectional area,
and there was a positive relationship between age and callosal T1 relaxation
times, suggestive of an increase in callosal water in 36 volunteers aged 26–79
years.40

Cerebellum

MRI studies
The studies investigating cerebellar volume have mostly suffered from very
small sample sizes, thereby failing to demonstrate decreases in cerebellar
volume with age.41 In a study of 69 healthy subjects aged 20–85, a subgroup
of a PET study showed significant decrements in the cerebellum.16 Larger
studies have shown some cerebellar atrophy, which is more pronounced in
males.25,42

Midbrain, pons, medulla

MRI studies
In a study of 36 normal volunteers aged 26–79, there was a highly significant
age-related decline in the cross-sectional area of the midbrain, less signifi-
cant decline in the anterior cerebellar vermis and no significant decline in
medulla, pons or fourth ventricle.43 The volume of the midbrain, as meas-
ured by anteroposterior and interpeduncular diameters, has been found to
decrease with age in an MRI study utilising stereological methods. 44 This
correlated with previous functional imaging of the nigrostriatal system. A
significant correlation of lacunar infarction increase with increasing age was
found in a small sample, with 12% of the lesions in the brain stem. 13 An
analysis from the Cardiovascular Health Study found that minimal changes
were evident in the brain stem in 15.3% and moderate-marked changes in
3.8%.18
58 THE AGEING BRAIN

Basal ganglia
MRI studies
Thirty healthy adult volunteers aged 20–80 years were studied with regard
to deep grey matter hypointensities, finding that the red nucleus, substantia
nigra and dentate nucleus were relatively unchanged.45 The globus pallidus
and the putamen showed hypointensities in the aged. 45 A significant cor-
relation of lacunar infarction increase with increasing age was found in a
small sample, with 54% of the lesions in the basal ganglia.13 In a study of 69
healthy subjects aged 20-85, a subgroup of a PET study, showed significant
decrements in brain matter volumes of the amygdala, thalamus, and cau-
date.16 A long term, five-year follow-up study of 24 successfully aged elderly
without significant neuropyschometric decline showed a modest increase in
white matter hyperintensities in the basal ganglia.17

Conclusion (Midline, Cerebellum, Midbrain, Basal Ganglia)

Involutional changes in midline, basal ganglia and infratentorial compart-
ments are evident, but have not been as well studied as cortical, ventricular
and gross brain changes. These studies particularly suffer from low numbers
of aged within the cross-sectional studies, thus possibly failing to have suf-
ficient power to detect the extent of change with ageing.

Imaging characteristics
MRI studies
Utilising quantitative MRI, a statistically significant relationship existed
between T1 (spin lattice relaxation time) in brain tissue in vivo and age for
10 brain structures investigated in 115 healthy controls aged 4–72.46 Least
squares regression analysis showed that T1 varied as a function of age in
the pulvinar nucleus, anterior thalamus, caudate, frontal white matter, optic
radiation, putamen, genu, occipital white matter and cortical grey matter. T1
declined throughout adolescence and early adulthood, reaching a minimum
in the fourth to sixth decade and then beginning to increase. These changes
differed in white matter such that the white matter reached minimum and
increased sooner than grey matter.46

Synthesis of findings
While there are many inconsistencies in the literature summarised above, the
following findings can be considered to be consistent features of brain ageing:
• Ventricular enlargement
• Reduction in gross brain volume
• Regional declines in frontal and temporo-parietal brain volume
• Increased cortical atrophy, especially in hippocampal and parahip-
pocampal regions
• Increased white matter hyperintensities, particularly in periventricu-
lar regions
STRUCTURAL NEUROIMAGING 59

• Some reduction in midline, basal ganglia and cerebellar volume

These changes occur earlier and progress more rapidly in males.

The validity and reliability of the above findings is limited by:

• Incomplete characterisation of “normal” aged persons
• Use of cross-sectional studies which may lack validity and have insuf-
ficient power to demonstrate significant change
• Variations in imaging technology
• Lack of standardisation of rating/measurements utilised to assess
change
• Lack of longitudinal studies of intra-individual change

Given the age-related changes noted, there is considerable evidence of measur-

able involutional change as assessed by structural neuroimaging.

Directions for future research

Whilst the majority of neuroimaging research demonstrates measurable
changes within the brain in normal ageing, refinement of investigative pro-
cedures would yield better data for comparison, especially with regard to
differentiation from neurodegenerative processes. We suggest the following
considerations for future studies:
• Better characterisation of the degree of normality of subjects by
physical examination, neuropsychological assessment and agreed
upon criteria.
• Larger sample sizes to increase power to detect change.
• Longitudinal as opposed to cross-sectional studies to assess intra-
individual change, which has more relevance than cross-sectional
studies of individuals from different age groups.
• Use of standardised methods of neuroimaging: slice selection, orien-
tation, parameters etc. Multi-centre collaboration may be necessary
for this to happen.
• Quantitative assessments of change to allow assessment of degree
and rate of change, as well as comparison between studies.
• Inclusion of midline, basal ganglia and infratentorial structures.
• The combining of functional and structural imaging to understand
the implications of the structural changes seen.
• The applications of newer technologies such as diffusion tensor
imaging, and diffusion and perfusion MRI to better understand the
brain changes of ageing.

Acknowledgements

The writing of this paper was supported in part by an NHMRC research

grant.
60 THE AGEING BRAIN

References

1. Drayer BP. Imaging of the aging brain. Radiology. 1988; 166:785–796.

2. Coffey CE. Anatomic imaging of the aging human brain. In: Coffey CE, Cummings
JL, editors. The American Psychiatric Press textbook of geriatric neuropsychiatry.
Washington, DC: American Psychiatric Press, 2000; 181–238.
3. Ylikoski A, Erkinjuntti T, Raininko R, Sarna S, Sulkava R, Tilvis R. White matter
hyperintensities on MRI in the neurologically non diseased elderly. Stroke. 1995;
26:1171–1177.
4. Yamaura H, Ito M, Kubota K, Matsuzawa T. Brain atrophy during aging: a
quantitative study with computed tomography. J Gerontol. 1980; 35:492–498.
5. Cala LA, Thickbroom GW, Black JL, Collins DWK, Mastaglia FL. Brain density
and cerbrospinal fluid space size: CT of normal volunteers. Am J Roentgenol.
1981; 2:41–47.
6. Zatz LM, Jernigan TL, Ahumada AJ. Changes on cranial computed tomography
with aging: intracranial fluid volume. Am J Neuroradiol. 1982; 3:1–11.
7. Takeda S, Matsuzawa T. Brain atrophy during aging: a quantitative study using
computed tomography. J Am Geriatr Soc. 1984; 32:520–524.
8. Takeda S, Matsuzawa T. Age-related brain atrophy: a study with computed tom-
ography. J Gerontol. 1985; 40:159–163.
9. Takeda S, Matsuzawa T. Age-related changes in volumes of the ventricles, cis-
ternae and sulci: a quantitative study using computed tomography. J Am Geriatr
Soc. 1985; 33:264–268.
10. Laffey PA, Peyster RG, Natahan R, Haskin ME, McGinley JA. Computed tomog-
raphy and aging: results in a normal elderly population. Neuroradiology. 1984;
26:273–278.
11. Stafford JP, Albert MS, Naeser MA, Sandor T, Garvey AJ. Age-related differences
in computed tomographic scan measurements. Arch Neurol. 1988; 45:409–415.
12. Coffey CE, Wilkinson WE, Parashos IA, Soady SA, Sullivan RJ, Patterson LJ,
Figiel GS, Webb MC, Spritzer CE, Djang WT. Quantitative cerebral anatomy of
the aging human brain; a cross-sectional study using magnetic resonance imaging.
Neurology. 1992; 42:527–536.
13. Matsubayashi K, Shimada K, Kawamoto A, Ozawa T. Incidental brain lesions
on magnetic resonance imaging and neurobehavioral functions in the apparently
healthy elderly. Stroke. 1992; 23:175–180.
14. Christiansen P, Larsson HBW, Thomsen C, Wieslander SB, Henriksen O. Age
dependent white matter lesions and brain volume changes in healthy volunteers.
Acta Radiol. 1994; 35:117–122.
15. Blatter DD, Bigler ED, Gale SD, Johnson SC, Anderson CV, Burnett BM, Parker
N, Kurth S, Horn SD. Quantitative volumetric analysis of brain MR: normative
database spanning 5 decades of life. Am J Neuroradiol. 1995; 16:241–251.
16. Murphy DG, DeCarli C, McIntosh AR, Daly E, Mentis MJ, Pietrini P, Szczepanik
J, Shapiro MB, Grady GL, Horwitz B, Rapoport SI. Sex differences in human
brain morphometry and metabolism; an in vivo quantitative magnetic resonance
imaging and positron emission tomography study on the effect of aging. Arch Gen
Psychiat. 1996; 53:585–594.
17. Wahlund LO, Almkvist O, Basun H, Julin P. MRI in successful aging, a five year
follow-up study form the eighth to the ninth decade of life. Mag Reson Imaging.
1996; 14:601–608.
18. Yue NC, Arnold AM, Longstreth WT Jr, Elster AD, Jungreis CA, O’Leary DH,
Poirier VC, Bryan RN. Sulcal, ventricular, and white matter changes at MR imag-
ing in the aging brain: data from the cardiovascular health study. Radiology.
1997; 202:33–39.
STRUCTURAL NEUROIMAGING 61

19. Coffey CE, Lucke JF, Saxton JA, Ratcliff G, Unitas LJ, Billig B, Bryan RN. Sex
differences in brain aging: a quantitaive magnetic resonance imaging study. Arch
Neurol. 1998; 55:169–179.
20. Courchesne E, Chisum HJ, Townsend J, Cowles A, Covington J, Egaas B, Har-
wood M, Hinds S, Press GA. Normal brain development and aging: quanitative
analysis at in vivo MR imaging in healthy volunteers. Radiology,. 2000; 216:
672–682.
21. Resnick SM, Goldszal AF, Davazikos C, Golski S, Kraut MA, Metter EJ, Bryan
RN, Zonderman AB. One-year age changes in MRI brain volumes in older adults.
Cereb Cortex. 2000; 10:464–472.
22. Hatazawa J, Ito M, Yamaura H, Matsuzawa T. Sex differences in brain atrophy
during aging: a quantitative study with computed tomography. J Am Geriatr Soc.
1982; 28:235–239.
23. Gomori JM, Steiner I, Melamed E, Cooper G. The assessment of changes in brain
volume using combined linear measurements — a CT-scan study. Neuroradiol-
ogy. 1984; 26:21–24.
24. Sandor T, Albert M, Stafford J, Kemper T. Symmetrical and asymmetrical changes
in brain tissue with age as measured on CT scans. Neurobiol Aging. 1990; 11:
21–27.
25. Xu J, Kobayashi S, Yamaguchi S, Iijima K, Okada K, Yamashita K. Gender
effects on age related changes in brain structure. Am J Neuroradiol. 2000; 21:
112–118.
26. Meyer JS, Takashima S, Terayama Y, Obara K, Muramatsu K, Weathers S. CT
changes associated with nromal aging of the human brain. J Neurol Sci. 1994;
123:200–208.
27. Golomb J, de Leon MJ, Kluger A, George AE, Tarshish C, Ferris SH. Hippocampal
atrophy in normal aging: an association with recent memory impairment. Arch
Neurol. 1993; 50:967–973.
28. Mu Q, Xie J, Wen Z, Weng Y, Shuyun Z. A quantitative MR study of the hippoc-
ampal formation, the amyggdala, and the temporal horn of the lateral ventricle
in healthy subjects 40 to 90 years of age. Am J Neuroradiol. 1999; 20:207–211.
29. Jack CR Jr, Petersen RC, Xu YC, Waring SC, O’Brien PC, Tangalos EG, Smith
GE, Ivnik RJ, Kokmen E. Medial temporal lobe atrophy on MRI in normal aging
and very mild Alzheimer’s disease. Neurology. 1997; 49:786–794.
30. Jack CR, Petersen RC, Xu YC, O’Brien PC, Smith GE, Ivnik RJ, Tangalos EG,
Kokmen E. Rate of medial temporal lobe atrophy in typical aging and Alzheimer’s
disease. Neurology. 1998; 51:993–999.
31. Jack CR, Petersen RC, Xu YC, O’Brien PC, Smith GE, Ivnik RJ, Boeve BF, Tan-
galos EG, Kokmen E. Rates of hippocampal atrophy correlate with change in
clinical status in aging and AD. Neurology. 2000; 55:484–489.
32. Magnotta VA, Andreason NA, Schultz SK, Harris G, Cizadlo T, Heckel D, Nop-
oulos P, Flaum M. Quantitative in vivo measurement of gyrification in the human
brain: changes associated with aging. Cereb Cortex. 1999; 9:151–160.
33. Zatz LM, Jernigan TL, Ahumada AJ. White matter changes in cerebral computed
tomography related to aging. J Comput Assist Tomo. 1982; 6:19–23.
34. Braffman BA, Zimmerman RA, Trojanowski JQ, Gonatas NK, Hickey WF,
Schlaepfer WW. Brain MR: patahlogic correlation with gross and histopathol-
ogy. 2. Hyperintense white-matter foci in the elderly. Am J Roentgenol. 1988;
151:559–566.
35. Salonen O, Autti T, Raininko R, Ylikoski A, Erkinjuntti T. MRI of the brain
in neurologically healthy middle-aged and elderly individuals. Neuroradiology.
1997; 39:537–545.
36. Longstreth WT Jr, Manolio TA, Arnold A, Burke GL, Bryan N, Jungreis CA,
Enright PL, O’Leary D, Fried L. Clinical correlates of white matter findings on
62 THE AGEING BRAIN

cranial magnetic resonance imaging of 3301 elderly people: the cardiovascular

health study. Stroke. 1996; 27:1274–1282.
37. de Leeuw F-E, de Groot JC, Achten E, Oudkerk M, Ramos LM, Heijboer R,
Hofman A, Jolles J, van Gijn J, Breteler MM. Prevalence of cerebral white matter
lesions in elderly people: a population based magnetic resonance imaging study.
The Rotterdam Scan Study. J Neurol Neurosur Ps. 2001; 70:9–14.
38. Schmidt R, Roob G, Kapeller P, Schmidt H, Berghold A, Lechner A, Fazekas F.
Longitudinal change of white matter abnormalities. J Neural Transm Supp. 2000;
59:9–14.
39. Hayakawa K, Konishi Y, Matsuda T, Kuriyama M, Konishi K, Yamashita K,
Okumura R, Hamanaka D. Development and aging of brain midline structures:
assessment with MR imaging. Radiology. 1989; 172:171–177.
40. Doraiswamy PM, Figiel GS, Husain MM, McDonald WM, Shah SA, Boyko OB,
Ellinwood EH Jr, Krishnan KR. Aging of the human corpus callosum: magnetic
resonance imaging in normal volunteers. J Neuropsych Clin N. 1991; 3:392–
397.
41. Escalona PR, McDonald WM, Doraiswamy PM, Boyko OB, Husian MM, Figiel
G, Laskowitz D, Ellinwood EH Jr, Krishnan KR. In vivo stereological assessment
of human cerebellar volume: effects of gender and age. Am J Neuroradiol. 1991;
12:927–929.
42. Raz N. Age and sex do not affect cerebellar volume in humans. Am J Neuroradiol.
1997; 18:594–595.
43. Shah SA, Doraiswamy PM, Husian MM, Figiel GS, Boyko OB, McDonald WM,
Ellinwood EH Jr, Krishnan KR. Assessment of posterior fossa structures with
midsagittal MRI: the effects of age. Neurobiol Aging. 1991; 12:371–374.
44. Doraiswamy PM, Na C, Husian MM, Figiel GS, McDonald WM, Ellinwood
EH Jr, Boyko OB, Krishnan KR. Morphometric changes of the human midbrain
with normal aging: MR and sterelogic findings. Am J Neuroradiol. 1992; 13:
383–386.
45. Milton WJ, Atlas SW, Lexa FJ, Mozley PD, Gur RS. Deep gray matter hypoin-
tensity patterns with aging in healthy adults: MR imaging at 1.5 T. Radiology.
1991; 181:715–719.
46. Cho S, Jones D, Reddick WE, Ogg RJ, Steen RG. Establishing norms for age-
related changes in proton T1 of human brain tissue in vivo. Magn Reson Imag.
1997; 15:1133–1143.
Chapter 5

NEUROPSYCHOLOGICAL,
SENSORY AND MOTOR
CHANGES WITH AGEING
Stephen R. Lord* and Rebecca St George

Introduction

In this chapter we review the studies that have addressed neuropsychological,

sensory and motor changes with ageing. Specifically, we examine documented
age changes in reaction time, working memory, selective attention, vision,
taste, smell, hearing, vestibular sense, proprioception, tactile sensitivity,
vibration sense and muscle strength. We then review age changes in compos-
ite factors such as postural control and gait stability that are underpinned by
many of these sensory, motor and integrative systems. Factors that contrib-
ute to the increased variability in functioning evident with increased age are
then considered, and health and lifestyle strategies for maximising functional
performance and independence in older age are discussed.

Part 1: Age-Changes

Neuropsychological factors

Reaction Time
A ubiquitous finding of ageing is the slowing of mental processes and behav-
iour.1 There is approximately a 25% increase in simple reaction time from
the twenties to the sixties, with further significant slowing beyond this age.

*To whom correspondence should be addressed.

64 THE AGEING BRAIN

Older people are slower in part because they are more careful and more
likely to sacrifice speed than accuracy, but carefulness does not explain
all the age-differences observed. Hertzog et al. 2 compared performance in
young and old subjects on a mental rotation task by varying instructions
to emphasize speed or accuracy. They found that older adults used a more
conservative strategy, preferring to sacrifice speed for accuracy but even
when this difference was taken into account, age differences remained. 2
Other factors such as amount of practice, the length of the preparatory
period, mode of response, health and motivational levels can also only
partly account for age changes.
In addition to the well-documented increases in reaction time with age
outlined above, studies that have examined the relationship between reac-
tion time and ageing involving complex motor tasks have found that there
are notable age-related increases in movement time in addition to age-related
increases in decision time.3 This is particularly the case for movements of
whole limbs and the whole body.4

Working memory and selective attention

There is substantial evidence that working memory declines in old age and
that this explains observed age decrements in complex cognitive tasks. 5,6
Salthouse7 has used working memory tasks involving simultaneous storage
and processing of information such as digit span backwards, computation
span and reading and listening span. This work shows that working memory
could account for about 50% of the association between age and tests of
fluid intelligence. Dual task experiments where one of the tasks involves
walking or balance and the other task involves working memory 8 or list
learning9 have shown a greater cost of dual tasks in older persons than in
young. Neuropsychological tests such as Trails B test have also been shown
to explain variance in a choice stepping reaction time task that mimics the
step requirement necessary to avoid a fall.10 These studies suggest that
there is an increase in the cognitive resources required for postural control
with age.

Sensory systems

Vision
As Chapter 6 reviews age-related declines in visual function, only a few key
issues are addressed here. The research into the decline of visual function is
extensive and it has been reported that ageing affects many vision processes.
These include visual acuity, contrast sensitivity, glare sensitivity, dark adap-
tation, accommodation and depth perception, especially beyond 40 years.11
Reduced vision has a significant impact on the behaviour and lifestyle of
older people. An older person with reduced vision is placed at an increased
risk of social isolation, postural imbalance and falling, and limited ability to
undertake daily activities.11,12
NEUROPSYCHOLOGICAL AND SENSORY CHANGES 65

Taste and smell

After the age of about 60 years people experience a gradual loss of taste and
smell. Taste thresholds increase as a result of taste buds diminishing in size
and number.13 Sweet and salty tastes are particularly affected leaving many
older people reporting that they no longer can taste or enjoy food. A related
decline in sense of smell has also been reported, with about 40% of people 80
years and older have difficulty identifying common substances by smell.14

Hearing
Gradual hearing loss begins at the age of 20 and becomes more accelerated
above the age of 70. Approximately one third of women over the age of 65
report a hearing difficulty,15 and it is estimated that by the age of 80, two
thirds of people will suffer a significant hearing loss.14 In particular, the abil-
ity to discriminate between tones of higher frequencies is reduced, as observed
in both cross-sectional16 and longitudinal studies.17 Comprehension of speech
becomes gradually more difficult due to increases in frequency discrimination
thresholds for the short tones present in human speech.18,19 Another aspect of
comprehending speech is auditory temporal discrimination and sound locali-
zation, factors that are also adversely affected by age.20,21 Older people also
experience difficulty in performing dichotic listening tasks, i.e. tasks requiring
them to attend to auditory information presented in one ear while ignoring
the information presented in the other.22

Vestibular sense
The vestibular system plays an important role in maintaining correct balance
and posture. Katsorkas23 studied over 1000 patients over the age of 70 and
found that over one third had a disease of vestibular origin. The vestibular
system contributes to posture by maintaining the reflex arc, keeping the head
and neck in the vertical position and by corrective movements elicited through
the vestibulo-ocular and vestibulo-spinal pathways.24 To preserve gaze during
movement of the head, a visual-vestibular interaction is required involving
the enhancement and suppression of compensatory eye movements (the ves-
tibulo-ocular reflex). Ageing has been found to be associated with diminished
ability to enhance and suppress the vestibulo-ocular reflex during horizontal
rotation.25 A number of researchers have found a reduced reactivity to caloric
and rotational stimulation in subjects over the age of 60.26,27 This decline in
vestibular function results in many elderly people experiencing dizziness and
unstable posture, increasing their likelihood of a fall.

Peripheral sensation
Receptors in the skin that respond to touch include Meissner and Pacinian
corpuscles, Merkel disks, Ruffini cylinders and free nerve endings.28 Pacin-
ian corpuscles are also sensitive to vibratory stimuli. Meissner and Pacinian
corpuscles in particular show considerable age-related losses in numbers and
morphological changes. Proprioception or kinaesthesis is served by special-
66 THE AGEING BRAIN

ised receptors in muscles, tendons, and joints to provide information on the

position and movements of body parts.29
Since the initial work by Pearson in 1928,30 age–related changes in vibra-
tion sense have been extensively studied. This work has consistently found age-
related declines in vibration sense to vibration frequencies greater than 50 Hz
on various parts of the body.31-37 It has also been found that vibration sense is
poorer in the lower limb compared with the upper limb at all ages and shows
a greater age-related decline.31-37 Compared with vibration sense, there have
been relatively few studies on the effect of age on tactile sensitivity. Like vibra-
tion sense, most reports indicate that tactile sensitivity, as measured by aesthe-
siometers or by two–point discrimination, decreases significantly with age37-41
and is reduced in the lower limb compared with the upper limb.37-39,41
Laidlaw and Hamilton were the first researchers to demonstrate an age-
related decline in joint position sense.42 They found that subjects aged 17
to 35 years had lower thresholds and superior ability to detect direction
of joint movements of the hip, knee and ankle than subjects aged 50 to 85
years. Since then, further studies have found significant age-related declines
in position sense of the knee joint, 43-46 metacarpophalangeal joint47 and
metatarsophalangeal joint.48 However, clinical studies that have investigated
whether there is a decline in joint position sense beyond 65 years of age have
produced inconsistent results. This may be due at least in part to the impreci-
sion of the tests used, which have been based on subjects’ ability to identify
experimenter-induced movements of body parts.48

Motor factors

Muscle strength
Numerous studies have reported a loss of both isometric and dynamic muscle
strength with increased age. In men, muscle strength appears to decrease only
marginally between 20 and 40 years, but beyond 40 years declines at an accel-
erated pace, so that hand grip strength is reduced by 16% and leg strength
by 28% in men aged 60–69 compared with men aged 20–29.49,50 In women,
muscle strength appears to decline from an earlier age and at a greater rate,
so that over the same age range handgrip strength declines by 20% and leg
strength by 38%.51 It has also been shown that muscle strength continues to
decline significantly beyond the sixties in both sexes.52 In studies that have
used both men and women, it has been found that muscle strength in women
is about 60-70% of that in men.49-51
Leg extensor power (the product of force and the rate of force genera-
tion) appears to decline at an even greater rate with age than does isometric
strength. In a cross-sectional study of 100 men and women aged 65–89
years, Skelton et al.53 found a loss of isometric strength of 1–2% per annum,
whereas the loss of leg extensor power was around 3.5 percent per annum.
Increased age is also associated with a deterioration of muscle elastic behav-
iour and reflex potentiation.54
NEUROPSYCHOLOGICAL AND SENSORY CHANGES 67

Reduced strength in the lower limbs has serious implications for older
people. Vandervoort and Hayes55 found impaired ankle plantarflexor muscle
force and power in residents of geriatric care facilities who were capable of
independently performing activities of daily living. Reduced strength is also
reflected in a difficulty in rising from a chair without the use of the hands,
and it has been found that an inability to undertake this task is associated
with subsequent disability, falls and fractures in older people.56–58

Postural stability

Standing
Normal standing is a dynamic process and characterized by small amounts
of postural sway. Sway is controlled by continual muscle activity (primarily
of the calf muscles) and requires an integrated reflex response to visual,
vestibular and somatosensory input59 that acts to inform the brain of the
position and movement of the body in three-dimensional space. The mus-
culoskeletal component of postural stability encompasses the biomechanical
properties of body segments, muscles and joints. Linking the sensory and
neuromuscular components are higher-level neurological processes that
enable anticipatory mechanisms for planning a movement, and adaptive
mechanisms for reacting in a controlled manner to the changing demands
of the particular task.60
The contribution of age-related declines in sensorimotor functioning to
increased postural sway in old age has been widely evaluated in the litera-
ture.61 Factors found to be correlated with increased sway include reduced
lower extremity calf muscle strength,62,63 reduced peripheral sensation,64,65
poor vision62,66 and slow reaction time.62,67 Smaller associations between
vestibular function and sway have been reported.62,68 Although the extent to
which one input can compensate for the loss of another is unclear, there is
some evidence that peripheral sensation is the most important sensory system
in the regulation of standing balance in older adults.59,62 Therefore, due to
its composite nature, standing instability is an indicator of sensory loss and
overall functional decline.

Gait
Many older people demonstrate a slow and laboured gait69,70 as a result of
shorter step length69–71 and more time spent in double limb support.69,70,72
Our group has found that older people with slow and variable gait demon-
strate reduced sensory acuity, lower limb muscle weakness, impaired vestibu-
lar function and slow reaction time. As with the sensorimotor factors that
underpin gait stability, slow and variable gait has been found to be a predictor
of falls and fractures in older people.69,73–76 Thus as with standing balance,
walking patterns provide an overall index of the extent of underlying sensory,
motor and central processing declines that occur with age.
 68 THE AGEING BRAIN

Part 2: Variability and Age Changes

Figure 1 presents findings from the Randwick Falls and Fractures Study40
that show the relationships between a range of sensorimotor factors and age.
Although it is apparent that there are considerable mean age-related declines,
it is evident that with increased age, there is also increased variability in
performance for each measure. For example, while there is an exponential
increase in reaction time with age, some older subjects perform similarly to
young subjects whereas others have scores indicating a two-fold slowing in
performance.
Understanding the causes of this variability may have important implica-
tions for health and quality of life in older people. Figure 2 shows a theo-
retical representation of the “normal” age-related decline in function of a
Quadriceps strength (kg)
Vestibular sense

Age group Age group

Vibration sense

Visual aculty

Age group Age group

Reaction time (ms)

Sway - foam

Age group Age group

Figure 1. Age—related changes in various sensory and motor functions—summary

findings of 550 women from the Randwick Falls and Fractures Study.40
The solid line shows the mean change with age and the grey band indicates
the inter-quartile range.
 NEUROPSYCHOLOGICAL AND SENSORY CHANGES 69
Functional level

Figure 2. Theoretical representation of the “normal” age-related decline in function

of a sensorimotor system that contributes to stability (grey-shaded area
represents the upper and lower bounds). The black line indicates the onset
of a disease such as a stroke which can rapidly change function and the
dashed line indicates a criterion level for adequate performance.

sensorimotor system that contributes to stability. The figure shows that up

until age 55 there is little change in function, but beyond this age there is a
progressive decline. This decline occurs in all people, however the variability
in function becomes progressively greater as age increases. Persons on the
lower band reach the criterion level for a critical loss of functioning (for
example a fall) by the age of 65 whereas those toward the upper band are
still above the criterion level at age 80. The figure also depicts a situation
in which the onset of significant disease such as a stroke can rapidly change
functional performance and result in performance levels below the criterion
level at any age.

Part 3: Intervention Strategies for Maximising Function

The role of exercise and good health habits

Good health habits have an impact on both the length of life and its quality
in old age. For example, it has been found that people who followed seven
healthy lifestyle habits: never smoked, moderate consumption of alcohol,
daily breakfast, no snacking, seven to eight hours of sleep per night, regular
exercise and ideal weight, lived on average nine years longer than those who
had none of the good health habits. Those with healthy habits also had
greater functional ability and suffered less from debilitating illnesses.77
The role of regular exercise, in particular is likely to be important for max-
imizing functional abilities, as age-related changes in many sensorimotor and
70 THE AGEING BRAIN

balance systems also occur with inactivity.14 Research has consistently shown
that older people who are active and highly conditioned are able to maintain
a higher functional ability, and are able to prolong their lifespan and delay
the onset and progression of chronic diseases14. Cross-sectional studies have
found that older people actively engaged in exercise perform better in a range
of sensori-motor function tests, including reaction time, strength, flexibility
and balance compared with matched groups of older non-exercisers.78
Furthermore, randomised controlled trials have now shown conclu-
sively that exercise can improve performance in these sensori-motor
parameters.79In a large, randomised, controlled trial of the effect of exer-
cise in elderly women, we found the exercisers showed significant improve-
ments in lower limb strength, simple reaction time, neuromuscular control,
standing and leaning balance and gait stability. 80–82 It appears that older
individuals adapt to resistive and endurance exercise training in a similar
fashion to younger people.14 This not only results in improved functional
ability but also beneficial effects on age-associated diseases such as Type II
diabetes, coronary heart disease, hypertension, osteoporosis, and obesity. 83
Thus, the development of exercise programs that are enjoyable and easily
accessible, so as to maximise long-term participation, may prove a valuable
strategy for maximizing sensorimotor functioning, mobility, independence
and quality of life in older people.

Conclusions

In summary, ageing is associated with significant declines in neuropsycho-

logical, sensorimotor and balance function. These declines are exponential
in nature, with noticeable changes evident in the 40s and 50s and functional
limitations common in those aged 80 years and over. Concomitant with
age-changes is an increase in variability in functional performance for most
measures. This variability may be due in part to genetic differences, although
factors such as disease and inactivity are likely to influence functional per-
formance. The maintenance of a “healthy lifestyle”, and in particular physical
activity, appears to be an important strategy for maximizing sensorimotor
functioning, mobility, independence and quality of life in older age.

References

1. Salthouse TA. Speed and age: multiple rates of age decline. Exp Aging Res. 1976;
2:349–359.
2. Hertzog C, Vernon MC, Rypma B. Age differences in mental rotation task
performance: the influence of speed/accuracy tradeoffs. J Gerontol. 1993; 48:
150–160.
3. Spirduso, WW. Reaction and movement time as a function of age and physical
activity level. J Gerontol. 1975; 30:435–40.
NEUROPSYCHOLOGICAL AND SENSORY CHANGES 71

4. Grabiner MD, Jahnigen DW. Modeling recovery from stumbles: preliminary data
on variable selection and classification efficacy. J Am Geriatr Soc. 1992; 40:
910–913.
5. Salthouse TA. Influence of processing speed on adult age differences in working
memory. Acta Psychol. 1992; 79:155–70.
6. Salthouse TA. Mechanisms of age-cognition relations in adulthood. Hillsdale:
Lawrence Erlbaum Ass, 1992.
7. Salthouse TA. Working-memory mediation of adult age differences in integrative
reasoning. Mem Cognition. 1992; 20:413–423.
8. Maylor EA, Wing AM. Age differences in postural stability are increased by
additional cognitive demands. J Gerontol. 1996; 51:143–54.
9. Lindenberger U, Marsiske M, Baltes PB. Memorizing while walking: increase
in dual-task costs from young adulthood to old age. Psychol Aging. 2000; 15:
417–36.
10. Lord SR, Fitzpatrick RD. Choice stepping reaction time: a composite measure of
falls risk in older people. J Gerontol. [In press].
11. Pitts DG. The effects of aging on selected visual functions: dark adaptation, visual
acuity, stereopsis, brightness contrast. In: Sekuler R, Kline DW, Dismukes K, edi-
tors. Aging in human visual functions. New York: Liss, 1982.
12. Lord SR, Dayhew J. Visual risk factors for falls. J Am Geriatr Soc. 2001; 49:
508–515.
13. Bartoshuk LM. Rifkin B. Marks LE. Bars P. Taste and aging. J Gerontol. 1986;
41(1):51–57.
14. Neiman DC. Fitness and sports medicine, 3rd ed. Mountain View, CA: Mayfield
Pub, 1995.
15. Ward JA, Lord SR, Williams P, Anstey K. Hearing impairment and hearing aid
use in women over 65 years of age. Med J Australia. 1993; 159.
16. Gates GA, Cooper JC, Kannel WB, Miller NJ. Hearing in the elderly: The Framin-
gam cohort, 1983-1985: Part I. Basic audiometric test results. Ear Hearing. 1990;
11:247–256.
17. Brant LJ, Fozard JL. Age changes in pure-tone hearing thresholds in a longitudinal
study of normal human aging. J Acoust Soc Am. 1990; 88:813–820.
18. Cranford JL, Stream RW. Discrimination of short duration tones by elderly sub-
jects. J Gerontol. 1991; 46:37–41.
19. Matschke RG. Frequency selectivity and psychoacoustic tuning curves in old age.
Acta Oto-Laryngol. 1990; 476 (Suppl.):114–119.
20. Schmitt JF, Carroll MR. Older listeners’ ability to comprehend speaker-generated
rate alteration of passages. J Speech Hear Res. 1985; 28:309–312.
21. Rastatter M, Watson M, Strauss-Simmons D. Effects of time-compression on
feature and frequency discrimination in aged listeners. Percept Motor Skill. 1989;
68:367–372.
22. Noffsinger D, Martinez CD, Andrews M. Dichotic listening to spech: VA-CD data
from elderly subjects. J Am Acad Audiol. 1996; 7:49–56.
23. Katsarkas A. Dizziness in ageing: a retrospective study of 1194 cases. Otolaryng
Head Neck. 1994; 110:296–301.
24. Stelmach GE, Worringham CJ. Sensorimotor deficits related to postural stability.
Implications for falling in the elderly. Clin Geriatr Med. 1985; 1:679–694.
25. Baloh RW, Jacobsen KM, Socotch TM. The effect of aging on visual-vestbulo-
ocular responses. Exp Brain Res. 1993; 95:509–16.
26. Karlsen EA, Hassanein RM, Goetzinger CP. The effects of age, sex, hearing
loss and water temperature on caloric nystagmus. Laryngoscope. 1981; 91:
620–627.
27. Ghosh P: Aging and auditory vestibular response. Ear Nose Throat J. 1985; 64:
264–266.
72 THE AGEING BRAIN

28. Kenshalo DR. Age changes in touch, vibration, temperature, kinesthesis and pain
sensitivity. In Birren JE, Schaie KW, editors. Handbook of the psychology of
aging. New York: Van Nostrand Reinhold, 1977.
29. Howard IP, Templeton WB. Human spatial orientation. London: John Wiley and
Sons, 1966.
30. Pearson GHJ. Effect of age on vibratory sensibility. Arch Neuro Psychiatr. 1928;
20:482–496.
31. Laidlaw RW, Hamilton MA. Thresholds of vibratory sensibility as determined by
the pallesthesiometer. Bull Neurol Inst NY. 1937; 6:494–503.
32. Cosh JA. Studies on the nature of vibration sense. Clin Sci. 1953; 12:131–151.
33. Mirsky IA, Futterman P, Broh-Kahn RH. The quantitative measurement of vibra-
tory perception in subjects with and without diabetes mellitus. J Lab Clin Med.
1953; 41:221–235.
34. Steiness IB. Vibratory perception in normal subjects. Acta Med Scandi. 1957; 158:
315–325.
35. Rosenberg G. Effect of age on peripheral vibratory perception. J Am Geriatr
Assoc. 1958; 6:471–481.
36. Perret E, Regli F. Age and the perceptual thresholds for vibratory stimuli. Eur
Neurol. 1970; 4:65–76.
37. Kenshalo DR. Somesthetic sensitivity in young and elderly humans. J Gerontol.
1986; 41:732–742.
38. Dyck PJ, Schultz PW, O’Brien PC. Quantification of touch-pressure sensation.
Arch Neurol. 1972; 26:465–473.
39. Bolton CF, Winkelmann RK, Dyck PJ. A quantitative study of Meissner’s corpus-
cles in man. Neurology. 1966; 16:1–9.
40. Lord SR, Ward JA. Age-associated differences in sensori-motor function and bal-
ance in community dwelling women. Age Ageing. 1994; 23:452–460.
41. Halar EM, Hammond MC, LaCava EC, Camann C, Ward J. Sensory perception
threshold measurement: an evaluation of semiobjective testing devices. Arch Phys
Med Rehab. 1987; 68:499–507.
42. Laidlaw RW, Hamilton NA. A study of thresholds in appreciation of passive
movement among normal control subjects. Bull Neurol Inst. 1937; 6:268–273.
43. Skinner HB, Barrack RL, Cook SD. Age-related decline in proprioception. Clin
Orthop Relat Res. 1984; 184:208–211.
44. Kaplan FS, Nixon JE, Reitz M, Rindfleish L, Tucker J. Age-related changes in
proprioception and sensation of joint position. Acta Orthop Scand. 1985; 56:
72–74.
45. Petrella RJ, Lattanzio PJ, Nelson MG. Effect of age and activity on knee joint
proprioception. Am J Phys Med Rehab. 1997; 76:235–241.
46. Hurley MV, Rees J, Newham DJ. Quadriceps function, proprioceptive acuity and
functional performance in healthy young, middle-aged and elderly subjects. Age
Ageing. 1998; 27:55–62.
47. Kokmen E, Bossemeyer RW, Williams WJ. Quantitative evaluation of joint
motion sensation in an aging population. J Gerontol. 1978; 33:62–67.
48. MacLennan WJ, Timothy JI, Hall MRP. Vibration sense, proprioception and
ankle reflexes in old age. J Clin Exp Gerontol. 1980; 2:159–171.
49. Petrovsky JS, Burse RL, Lind AR. Comparison of physiological responses of men
and women to isometric exercise. J Appl Physiol. 1975; 38:863–868.
50. Murray MP, Gardner GM, Mollinger LA, Sepic SB. Strength of isometric and
isokinetic contractions. Knee muscles of men aged 20 to 86. Phys Ther. 1980;
60:412–419.
51. Murray MP, Duthie EH, Gambert SR, Sepic SB, Mollinger LA. Age related
differences in knee muscle strength in normal women. J Gerontol. 1985; 40:
275–280.
NEUROPSYCHOLOGICAL AND SENSORY CHANGES 73

52. MacLennan WJ, Hall MRP, Timothy JI, Robinson M. Is weakness in old age due
to muscle wasting ? Age Ageing. 1980; 9:188–192.
53. Skelton DA, Greig CA, Davies JM, Young A. Strength, power and related
functional ability of healthy people aged 65-89 years. Age Ageing. 1994; 23:
371–377.
54. Bosco C, Komi PV. Influence of aging on the mechanical behaviour of leg extensor
muscles. Eur J Appl Physiol. 1980; 45:209–219.
55. Vandervoort AA, Hayes KC. Plantarflexor muscle function in young and elderly
women. Eur J Appl Physiol. 1989; 58:389–394.
56. Nevitt M, Cummings S, Kidd S, Black D. Risk factors for recurrent nonsyncopal
falls. J Amer Med Assoc. 1989; 261:2663–2668.
57. Campbell AJ, Borrie MJ, Spears GF. Risk factors for falls in a community-based
prospective study of people 70 years and older. J Gerontol. 1989; 44:M112–
117.
58. Lipsitz LA, Jonsson PV, Kelley MM, Koestner JS. Causes and correlates of recur-
rent falls in ambulatory frail elderly. J Gerontol. 1991; 46:M114–122.
59. Fitzpatrick R, Rogers DK, McCloskey DI. Stable human standing with lower-
limb muscle afferents providing the only sensory input. J Physiol. London, 1994;
480(2): 395–403.
60. Shumway-Cook A, Woollacott M. Motor control: Theory and practical applica-
tions. Baltimore: Williams and Wilkins, 1995.
61. Lord SR, Sherrington C, Menz HB. Falls in older people: Risk factors and strate-
gies for prevention. Cambridge: Cambridge University Press, 2001.
62. Lord SR, Clark RD, Webster IW. Postural stability and associated physiological
factors in a population of aged persons. J Gerontol. 1991; 46(3):M69–76.
63. Satariano WA, DeLorenze GN, Reed D, Schneider EL. Imbalance in an older
population: an epidemiological analysis. J Aging Health. 1996; 8(3):334–358.
64. MacLennan WJ, Timothy JI, Hall MRP. Vibration sense, proprioception and
ankle reflexes in old age. J Clin Exp Gerontol. 1980; 2: 159–171.
65. Duncan G, Wilson JA, MacLennan WJ, Lewis S. Clinical correlates of sway in
elderly people living at home. Gerontology. 1992; 38: 160–166.
66. Lichtenstein MJ, Shields SL, Shiavi RG, Burger MC. Clinical determinants of
biomechanis platform measures of balance in aged women. J Am Geriatr Soc.
1988; 36:996–1002.
67. Stelmach G, Phillips J, DiFabio R, Teasdale N. Age, functional postural reflexes,
and voluntary sway. J Gerontol. 1989; 44(4):B100–106.
68. Cohen H, Heaton LG, Congdon SL, Jenkins HA. Changes in sensory organisation
test scores with age. Age Ageing. 1996; 25:39–44.
69. Lord SR, Lloyd DG, Li SK. Sensori-motor function, gait patterns and falls in
community dwelling women. Age Ageing. 1996; 25:292–299.
70. Finley FR, Cody KA, Finizie RV. Locomotion patterns in elderly women. Arch
Phys Med Rehab. 1969; 50:140–146.
71. Bohannon RW. Comfortable and maximum walking speed of adults aged 20-79
years: reference values and determinants. Age Ageing. 1997; 26:15–19.
72. Winter DA, Patla AE, Frank JS, Walt SE. Biomechanical walking patterns in the
fit and healthy elderly. Phys Ther,. 1990; 70(6):340–347.
73. Woolley SM, Czaja SJ, Drury CG. An assessment of falls in elderly men and
women. J Gerontol. 1997; 52A(2):M80–87.
74. Cho C-Y, Kamen G. Detecting balance deficits in frequent fallers using clinical
and quantitative evaluation tools. J Am Ger Soc. 1998; 46:426–430.
75. Wolfson L, Whipple R, Amerman P, Tobin JN. Gait assessment in the elderly:
a gait abnormality rating scale and its relation to falls. J Gerontol. 1990; 45(1):
M12–19.
74 THE AGEING BRAIN

76. Luukinen H, Koski K, Laippala P, Kivela S-L. Risk factors for recurrent falls in
the elderly in long-term institutional care. Public Health. 1995; 109:57–65.
77. Breslow L, Breslow N. Health practices and disability: Some evidence from
Alameda county. Pre Med. 1993; 22:86–95.
78. Wagner EH, LaCroix AZ, Buchner DM, Larson EB. Effects of physical activity
on health status in older adults. I: Observational studies. Annu Rev Publ Health.
1992; 13:451–68.
79. Buchner DM, Beresford SA, Larson EB, La Croix AZ, Wagner EH. Effects of
physical activity on health status in older adults II: Intervention studies. Annu
Rev Publ Health. 1992; 13:469–488.
80. Lord SR, Ward JA, Williams P, Strudwick M. The effect of a 12 month exercise
program on balance, strength and falls in older women: a randomised controlled
trial. J Am Geriatr Soc. 1995; 43:1198–1206.
81. Lord SR, Ward JA, Williams P. The effect of exercise on dynamic stability in
older women: a randomised controlled trial. Arch Phys Med Rehab. 1996; 77:
232–236.
82. Lord SR, Lloyd DG, Nirui M, Raymond J, Williams P, Stewart RA. The effect of
exercise on gait patterns in older women: a randomised controlled trial. J Geron-
tol. 1996; 51A:M64–M70.
83. Rogers MA, Evans WJ. Changes in skeletal muscle with aging: Effects of exercise
training. Exercise Sport Sci R. 1993; 21:65–102.
Chapter 6

COGNITIVE CHANGES
AND THE AGEING BRAIN
Helen Christensen* and Rajeev Kumar

Introduction

There is still debate about the specific types of changes in cognitive and
intellectual functioning that occur over the lifespan, although there is also
agreement that cognitive change is not unitary and that some abilities decline
more rapidly than others.1–3 There is considerable interest and speculation
about the age of onset of cognitive deterioration and the brain mechanisms
responsible for it. More recently there has been greater appreciation of the
way in which individuals differ in their rates of cognitive change (i.e., inter-
individual variability), and in the possible importance of inconsistency in
cognitive performance (intra-individual variability) as a predictor of cogni-
tive deterioration. Two areas of intense investigation in cognitive ageing at
the present time include the hypothesis that there is a common factor that is
responsible for changes in both cognitive abilities and non-cognitive variables
over the lifespan, and the question of which brain structures are associated
with cognitive change.
This chapter summarizes recent evidence on the nature of cognitive
decline, the variability in individual responses to ageing and on the newer
research questions in cognitive ageing. Many previous reviews describe the
relationship between cognitive performance and brain structures from a neu-
ropsychological perspective rather than the individual differences perspective
taken in the present chapter. The neuropsychological approach emphasizes
the relationship between cognitive functions and brain locations and sys-
tems, and is informed by clinical neuropsychological methods, where lesions

*To whom correspondence should be addressed.

76 THE AGEING BRAIN

and patient behaviour are examined. An example of a neuropsychological

approach to ageing is the investigation of the frontal lobe hypothesis. This
hypothesis states changes in cognitive functioning are due to shrinkage or
other changes in the frontal lobes, and hence deficits should be specific to
tests which reflect frontal rather than other lobe functioning.4 In contrast to
this approach, the individual differences approach of this chapter examines
cognitive domains that have been derived from psychometric studies of the
structure of intelligence.
The chapter is divided into three sections. First, data are summarized to
demonstrate that cognitive ageing is not unitary but rather some abilities
decline more rapidly than others. Second, the evidence is examined which
suggests that there is greater variability in test scores as a function of age, a
finding which may indicate the existence of a number of sub-groups which
age at different rates within the same population group. Finally, the evidence
is examined for the common cause hypothesis and for the association of vari-
ous brain structures with change in major areas of cognitive function.

Constraints, limitations and definitions

In this chapter, cognitive ageing refers to the cognitive performance that is
observed in longitudinal and cross-sectional studies of community-dwelling
elderly people. Methodologically, longitudinal studies provide information
that cross-sectional studies cannot provide such as: estimates of individual
rates of decline, risk factors for such decline, data on correlations between
changes in cognitive ability and changes in other cognitive and non-cogni-
tive domains. However, longitudinal studies underestimate change because
of practice effects and selective attrition. In particular, individuals who later
go on to develop dementia are often excluded from community longitudinal
studies since they no longer live in the community, and individuals who die
before follow-up are known to have lower scores at baseline. Conversely,
cross-sectional studies tend to over-estimate cognitive decline. One reason
is that there are established cohort effects in intelligence, with an average
increase in intelligence being reported for successive generations of individu-
als. Flynn5 concluded that the average IQ had increased by 14 points (corre-
sponding to approximately one standard deviation of intelligence measures)
from the period 1932 to 1978.
Although there are many methods used to describe or examine cognitive
change we characterize cognitive ability as consisting of three major abili-
ties: crystallized intelligence, memory and cognitive speed. The distinction
between crystallized intelligence and cognitive speed or fluid intelligence
was not invoked to describe the nature of decline in intellectual abilities, but
arose from factor analytic studies of the structure of intelligence.6 Crystal-
lized intelligence is “assumed to be the cumulative end-product of informa-
tion acquired” by an individual7 and is demonstrated on tests of vocabulary,
information accumulation and other knowledge-based activities. Memory is
commonly divided into short-term and long-term, with both types further
COGNITIVE CHANGES 77

Box 1. Common measures of intelligence and memory

Ability type Common measures and tests

Crystallized intelligence Vocabulary

Information
Similarities
National Adult Reading Test
Wechsler Adult Intelligence Scale (Verbal Scale)
Fluid intelligence Raven’s Progressive Matrices
Block Design
Digit Symbol Substitution
Reaction Time
Wechsler Adult Intelligence Scale (Performance Scale)
Procedural memory Priming
Reversed mirror reading
Classical conditioning
Declarative memory Logical Memory and Paired Associates
Buschke Selective Reminding Test

fractionated into declarative and procedural memory. Declarative memory

refers to conscious recollection and recall, while procedural memory (which
is memory that does not require the intentional or conscious recollection of
an experience) includes priming, classical conditioning and skill-based learn-
ing. Cognitive speed refers to performance on perceptual-motor tasks which
are timed, and overlaps with more traditional constructs such as fluid intel-
ligence. Examples of such tasks are the Digit Symbol Substitution test from
the Wechsler Adult Intelligence Scale8 or simple and choice reaction time
tasks. There is enormous debate as to the nature of speed and speed tasks,
and whether it is in fact possible to develop a core speed task.9 Box 1 shows
typical cognitive tests and their assignment to these key concepts.

Cognitive Change is not Unitary

The existence of different developmental trajectories according to ability

type has been well recognized for decades.6,10 Data from meta-analyses of
cross-sectional studies7 suggest that crystallized intelligence remains relatively
stable in old age, dropping in late old age. In contrast, cognitive speed drops
by approximately 20% at age 40 and by 40–60% at age 80. Other large
cross-sectional data sets, including those from the Wechsler Adult Intelli-
gence Scale-Revised (WAIS-R) have been summarized. 11 Results from the
WAIS performance measures, such as Digit Symbol Substitution and Block
Design, show steep gradients while WAIS-R Verbal tests such as Information,
78 THE AGEING BRAIN

Vocabulary, and Comprehension, indicate shallower declines across the age

range examined (20–74 years).
Data from many longitudinal studies confirm these findings (see Box 2 for
a summary of longitudinal studies worldwide examining cognitive change).
Figure 1 illustrates changes in crystallized intelligence (vocabulary), percep-
tual speed (identical pictures), and memory (immediate recall) using data that
have been derived from the Seattle Longitudinal Study, which is one of the
most influential longitudinal studies ever undertaken. This figure, adapted
from Schaie2 illustrates that changes in speed and verbal memory appear
to accelerate after 60 years of age, while verbal ability is more robust and
may only drop in very late old age. Participants from the study were drawn
from an age/sex stratification of members of a health maintenance organiza-
tion. Schaie2 reported that cohort effects were evident for the baby boomer
generation, with deterioration abating in these newer age cohorts. More
specific data on age changes after the age of 70 years are illustrated in com-
munity-based surveys of older samples, such as the sample reported by us in
the Canberra Longitudinal Study12 (Figure 2). These data clearly show that
crystallized intelligence does not change significantly, even into late old age,
in those surviving to follow-up.
The findings from our study showing significant deterioration in memory
performance have been reported in many other longitudinal data sets: the
Einstein ageing study13, Victorian Longitudinal Study,3 Duke Longitudinal
Study and other Australian longitudinal studies (Australian Longitudinal
Study of Aging), to name just a few. Similar results have been reported for
processing speed.2,13 There is some evidence that the rate of ageing in these
cognitive domains may accelerate once individuals reach 70, although this is
as yet unclear.
Cognitive decline is a predictor of mortality14 with cognitive test scores
on many measures predicting mortality even when demographic and health
variables have been controlled statistically. 15,16 While sensory, speed and
cognitive tests predict mortality, performance on certain tests of pre-morbid
ability such as National Adult Reading Test17 is not a predictor.

Inter-Individual and Intra-Individual Variability associated with Ageing

Older age is associated with greater inter-individual and intra-individual dif-

ferences. Thus, there is greater diversity in the cognitive trajectories of older
individuals (a between-individuals effect) and greater inconsistency in indi-
vidual responses to task performance (a within-individuals effect).

Inter-individual differences (diversity)

Although the average performance on most cognitive tasks declines with age,
studies have suggested that many older individuals change very little, whereas
others deteriorate dramatically. At older ages there is a greater diversity of
COGNITIVE CHANGES 79

Box 2: Longitudinal Studies of Psychological Ageing

Current Longitudinal Studies

Australian Longitudinal Study of Aging (ALSA; 1992– )
Baltimore Longitudinal Study of Aging (BLSA; 1958– )
Berkeley Older Generation Study
Berlin Aging Study (BASE; 1990– )
The Betula Project (1988– )
Canberra Longitudinal Study (1991– )
Einstein Aging Studies
Gender Study of Unlike-Sex DZ Twins (GENDER)
Georgia Centenarian Study (1988– )
Groningen Longitudinal Aging Study (GLAS; 1993– )
The Gerontological and Geriatric Population Studies in Göteborg, Sweden (H-70;
1971– )
Health and Retirement Study (HRS; 1992– ) and Asset and Health Dynamics
Among the Oldest Old (AHEAD; 1993– )
Interdisciplinary Longitudinal Study of Adult Development (ILSE; 1996– )
Italian Longitudinal Study of Ageing (ILSA)
The Kungsholmen Project (1987– )
Long Beach Longitudinal Study (1978– )
Longitudinal Aging Study Amsterdam (LASA; 1991– )
Lund 80+ Study (1988– )
Maastricht Aging Study (MAAS; 1992– )
Manchester and Newcastle Longitudinal Studies of Aging
McArthur Successful Aging Studies
Mills Longitudinal Study of Women (1956– )
National Growth and Change Study (NGCS)
New England Centenarian Study
Normative Aging Study (1963– )
Nordic Research on Aging (NORA; 1989– )
The Nun Study (1986– )
Octogenarian Twin Study (OCTO-Twin; 1990– )
San Antonio Longitudinal Study of Aging (SALSA)
Seattle Longitudinal Study (1956– )
The Swedish Adoption/Twin Study of Aging (SATSA; 1984– )
The Victoria Longitudinal Study (VLS; 1986– )
WAIS Longitudinal Study Database (NGCS 2000)

Completed Longitudinal Studies

AT&T Longitudinal Studies of Managers
The Bonn Longitudinal Study (BOLSA; 1965–1984, mortality follow-up)
First Duke Longitudinal Study (1955–1976 )
Second Duke Longitudinal Study (1968–1976)
Intergenerational Growth Study (IGS; 1932–1982)
New York State Psychiatric Institute Study of Aging Twins (1946–1973)
New York Longitudinal Study (1956–1988)
NIMH Study (1955)
Iowa State Study (1919–1976)
Terman-Stanford Study ()

Source: http://www.personal.psu.edu/faculty/s/m/smh21/index.htm76
80 THE AGEING BRAIN

Figure 1. Age changes in cognitive test scores over age range of 25 to 88 based on
two data points (1984 and 1991). Each age segment is based on a single
sample followed over seven years. T-score points are scaled to have a mean
of 50 and a standard deviation of 10. Data are taken from Schaie.2

(a)
COGNITIVE CHANGES 81

(b)

(c)

Figures 2. Mean scores on measures of crystallized intelligence (a), memory (b), and
speed (c) for four age groups over three occasions from the Canberra Lon-
gitudinal Study.
82 THE AGEING BRAIN

cognitive scores.18–21 In a review of six longitudinal and 48 cross-sectional

studies, Nelson and Dannefer18 reported increased variability with age, not
only for cognitive variables but also for personality and biological indices.
There have also been recent studies reporting greater individual differences
in cognitive change,3,12 findings that indicate greater diversity in the rates of
change at older ages. Most, but not all studies, suggest that the individual
differences in speed and memory increase with increasing age. These findings
are consistent with the possibility that a number of sub-groups exist with
different rates of cognitive change.
Data illustrating increased individual differences in change from the Can-
berra Longitudinal Study12 are displayed in Figure 3. Changes on a verbal
measure (crystallized intelligence) and a measure of memory are shown. These
data illustrate individual paths for all participants over the three occasions of
measurement. Measures were taken in 1991, 1994 and 1998. A measure of
individual diversity was calculated by regressing cognitive change scores on
age, determining the absolute residuals of all change scores and then correlat-
ing these residual scores with age. We found a positive correlation between
residuals scores and age both for speed and memory, indicating greater diver-
sity in these cognitive abilities at older ages. This association was not found
for crystallized intelligence.
Many factors have been suggested as sources of these individual differences
in cognitive trajectories. Following Kraemer et al.22 these can be conveniently
identified as either marker variables or risk factor variables. Marker variables
are those that are essentially unchangeable, such as education (schooling),
ApoE ε4 allele and gender. Risk factor variables, such as physical activity and
age, are those that develop and change with the individual. Because there are
data from cross-sectional studies on the risk factors of education, ApoE (a
marker variable) and health these are examined below.
Education has often been reported to be a protective factor in cognitive
ageing with a number of studies reporting that the rate of decline is less rapid
in the highly educated.23–25 Other studies report that the rates do not differ
as a function of educational,3 or that the effects are restricted to sub-groups
or only observed on verbal intelligence measures.26 To account for the asso-
ciation between education and the rate of cognitive decline, a number of
hypotheses have been advanced. Education has been suggested to be a proxy
variable for “brain reserve”, i.e. increased brain capacity across the lifespan
or pre-morbid intelligence (see Chapter 13). Low intelligence has been found
to predict decline on memory tests.27 There is evidence23 that the effect of
education on levels of performance may depend on the type of cognitive
outcome measure that is investigated. For example, the findings of a recent
review23 were that studies that measured outcome with a screening test, such
as the MMSE, reported a protective effect of education. Five of seven studies
that used a memory test reported faster decline in those with poor education,
and three of four studies found faster decline for measures of crystallized
intelligence. In comparison, three studies that used a measure of cognitive
COGNITIVE CHANGES 83

(a)

(b)

Figure 3. Longitudinal individual trajectories for (a) crystallized intelligence and (b)
memory from the Canberra Longitudinal Study.
84 THE AGEING BRAIN

speed failed to find an effect for education. These findings suggest that the
protective effects of higher education may be restricted to crystallized intel-
ligence, memory and mental state measures.
The relationship between cognitive change and ApoE ε4 and ApoE ε2
has also been investigated in a number of studies. With respect to ApoE
ε4, Anstey and Christensen23 reported that ApoE ε4 predicted cognitive
decline in five studies,28–32 three studies found that the effect was present
but observed only on certain tests33–35 and two studies reported that the rate
of change did not differ.36,37 Three more studies have recently been reported.
One study indicated faster decline in ApoE ε4 individuals,38 one suggested the
association was only present in those with MMSE below a score of 2739 and
one demonstrated no association in a very old sample.40 Overall, these find-
ings support the view that the rate of cognitive decline is greater in those with
ApoE ε4. The effect of ApoE ε4 was found primarily on measures of memory,
MMSE and processing speed, suggesting the effect is related to memory and
fluid abilities. Only one study reported results for a measure of crystallized
intelligence and this was non-significant.33
There is a consensus amongst researchers that serious health problems may
be detrimental to cognitive functioning but that less serious and self-reported
health problems do not account for the bulk of cognitive ageing. To quote
Arbuckle et al.26

“Cross-sectional and longitudinal studies of the relation between

health and cognitive functioning have generally shown that poorer
physical health is associated with poorer cognitive functioning …
but that age-associated illnesses explain only part of the age-related
variance in cognition.”

Anstey and Christensen23 summarized findings from eight longitudinal studies

reporting data on health and disease measures as predictors of cognitive
change. The measures of health included self-ratings, disease status, lung
function and depressive symptoms. Using a global self-report measure, three
studies found no relationship between self-reported poor health and cognitive
change, while two studies found relationships in the predicted direction (low-
ratings were associated with greater change). Objective measures of health
status including lung function and glucose tolerance were all related to
cognitive decline as was atrial fibrillation and cardiovascular disease. These
findings suggest that poor health is associated with faster cognitive decline.
However, as noted above, poor health may account for only a part of the
deterioration. Salthouse et al.41 have conduced analyses investigating the role
of health by comparing the magnitude of age-related effects on measures of
functioning, mostly speed, before and after statistical control of measures of
health. Between 15% and 20% of the age-related variance in cognitive scores
was accounted for by self-rated health, suggesting that health factors do not
mediate the great proportion of age-related cognitive decline.
COGNITIVE CHANGES 85

Although there are relatively few studies for each predictor, these findings
suggest that higher education, absence of the ApoE ε4 allele, and good health
are protective of subsequent cognitive decline. An interesting finding is that
risk factors may be specific to the type of cognitive task, with, for example,
education exerting an effect on crystallized and memory measures, and ApoE
ε4 allele affecting memory and speed to a greater extent than crystallized
intelligence.

Intra-individual differences (short-term and long-term individual difference)

Greater intra-individual change with older age has also been reported. Nes-
selroade42 made a distinction between individual change that is more or less
durable or systematic (developmental change) and individual change that is
transient and short term (the “wobble” about the developmental change).
Both these types of within-person change have been examined in a variety
of studies.43-45 In the original research of this kind, Hertzog et al46 tested
memory for sentences fortnightly in seven women over two years, testing
performance on equivalent forms of test. Individuals varied in the extent to
which they were consistent in their level of remembering. Individuals also had
different trajectories, with either steady improvement or consistent decline
over the two-year period. Recent studies of intra-individual variability show
that the wobble increases with age for memory,44 reaction time,43 and sen-
sori-motor tasks.44 The extent of deviation from the average is also a predic-
tor of poorer levels of performance.45 Li and Lindenberger44 have suggested
that transient intra-individual variability indicates impaired neurobiological
functioning. This variability seems to occur across different tasks (such as
sensori-motor and reaction time) and, hence, may be a relatively stable char-
acteristic of people. Individuals diagnosed with dementia show approximately
twice as much wobble as those not so diagnosed.46

The Common Cause Hypothesis

One of the major issues concerning cognitive researchers has been whether
cognitive decline in various cognitive domains (for example memory and fluid
intelligence) have independent rates of decline and independent predictors of
decline. In a famous paper,20 Patrick Rabbitt asked “Does it all go together
when it goes?”, and recent findings suggest that there may not be independ-
ence in the rate of change. Speed of processing (i.e. the rate at which indi-
viduals could process incoming stimuli) was suggested by many as a possible
common mechanism. Evidence from a number of sources47 indicated that
much of the age-related variance among speeded tasks and other cognitive
tasks was shared. In an influential development, Lindenberger and Baltes48
reported that sensory functioning, measured by vision and hearing assess-
ments, also shared age-related variance with cognitive tasks, suggesting that
a common mechanism might need to account for sensory changes. Even more
86 THE AGEING BRAIN

interesting have been reports indicating that grip strength and lung capacity
correlate with changes in cognitive functioning.43 These observations have
given rise to the “common cause” hypothesis48 — the hypothesis that a single
common mechanism may be responsible for the decline in physical, sensory
and cognitive processes.
Evidence for the common cause hypothesis in cognitive ageing relies on a
number of observations made from cross-sectional studies of age-heterogene-
ous samples. These include the observation that “cognitive primitives”, such
as sensory functioning and speed of processing, and non-cognitive factors
such as respiratory efficiency, low blood pressure and physical strength are
significantly intercorrelated in cross-sectional samples and decline conjointly
with age. A second observation is that age has only a small direct role in
predicting cognitive performance over its effect via sensory functioning or
speed of processing.14,47,48 That is, memory performance can be accounted
for almost entirely by performance on speed or sensory tasks, and that the
additional effect of age as a variable predicting cognitive performance once
these other variables are considered is minimal. A number of studies have
indicated that a common factor can be modelled in cross-sectional data sets.
For example, Salthouse et al.47 estimated a common factor directly, with the
effect of age being regressed onto both the factor and on to its indicators.
Fitting this model to five datasets, they found that speed, cognitive function-
ing, grip strength, and visual tests loaded on a common factor. Age had an
(additional) independent relationship with sensory functioning but not with
the other indicators of the common factor.
To explain the large proportion of shared age-related variance, research-
ers have proposed that a common physiological process is or processes are
responsible, for example, the “ageing brain”,48 changes in the central nervous
system and general fitness of the organism,49 and the ageing-related changes
to the physiology of the entire organism.50 More recently, Salthouse and
Czaja51 have suggested that the common cause may be of two types, which
may not be mutually exclusive. The first type of broad influence is cognitive,
either a particular type of process or a property of processing, such as speed
of processing or the involvement of working memory. A second is neuro-
anatomical or neurophysiological (either an area of processing, such as the
dorso-lateral prefrontal cortex) or a specific process (such as the dopaminer-
gic system). Salthouse and Czaja51 note that

“... the results of … analyses impose clear constraints on the nature

of plausible explanations for cognitive aging phenomena. That is,
because many age-related effects seem to operate at relatively broad
levels, which affect a wide variety of cognitive variables, researchers
must apparently postulate some general or nonspecific explanatory
mechanisms.”
COGNITIVE CHANGES 87

Data from cross-sectional research43,52 showing associations among cognitive

scores and other non-cognitive variables such as grip strength and respiratory
function suggest that the common process may be even more broad than brain
processes as identified by Salthouse and Czaja.51
The common cause hypothesis needs evaluation in longitudinal studies
where the associations among longitudinal changes in sensory, physical and
cognitive processes can be evaluated for their independence. To advance this
area, researchers need to examine longitudinal relationships among biologi-
cal, cognitive and sensory indicators preferably in large samples to maximize
power. As noted by Deary,9 “There is lack of tractable measures of the brain’s
information processing capabilities that will bridge behaviour and biology
with an explanation.”
In the following section, we review the association between MRI indices
of brain structure and cognitive performance. This represents one immediate
area of investigation that could help to understand what mediates changes in
cognitive performance across the lifespan.

Longitudinal Studies of MRI, CT and Cognitive Change

Tables 1 and 2 outline longitudinal studies53–74 examining the association

between various areas of the brain (as shown by MRI and CT scans) and meas-
ures of cognitive performance. Three groups of individuals were examined:
those with Alzheimer’s disease (AD) or probable AD, those with Cognitive
Decline, Mild Cognitive Impairment (MCI), or described as being “at risk”
and those described as “normal”. Information is summarized in these tables
about the number of participants in each of the studies, the length of the study
follow-up, the age of the participants, the brain area under study, whether
the study found significant changes in brain area volumes and whether the
study reported change in cognitive performance across a range of measures.
If significant volume change or cognitive change was found in each study, a
“Yes” was entered into the appropriate column. The final column of each
of the tables summarizes the findings from those studies where the correla-
tion between a cognitive measure and a brain volume measure was examined
explicitly. A “Yes” in this column indicates that a positive association was
found. A “No” indicates that the correlation was not significant.
The MRI studies (Table 1) suggest that over time there are changes in brain
volumes for normal individuals, those at risk of AD, those with MCI and for
those with AD. The degree of volume loss is greater in those with AD than
in those with other conditions59,62 and for those at older ages.58 The longi-
tudinal evidence also suggests that cognitive change is particularly evident
for individuals who were diagnosed with MCI and AD at baseline. Over the
range of follow-ups, normal individuals also show cognitive change, a finding
that is consistent with evidence presented earlier in this chapter. At shorter
follow-up intervals, there is less clear evidence for cognitive decline in normal
88 THE AGEING BRAIN

Table 1. Longitudinal studies reporting MRI changes and their associations with
cognitive change.

Study Number Length Age Brain Change Change Correlationb

of of (years) areaa in brain in
subjects study area cognitionb
(years)

Alzheimer’s Disease (AD) or Probable AD

Fox 199654 11 1 53.8 B Yes ? ?
Jack 199858 24 1 80.42 H Yes ? ?
Fox 199959 29 6 58.1 B Yes Yes Yes
Fox 200062 18 1 65 B Yes Yes ?
Jack 200064 28 3 73.8 H Yes Yes ?

Cognitive decline, Mild Cognitive Impairment (MCI), or “at risk”

Kaye 199755 12 3.6 84 B,H,TL Yes Yes Yes
Fox 199857 63 6 44.7 H Yes ? ?
Jack 200064 43 3 77 H Yes Yes ?

Normal individuals
Wahlund 199653 24 5 79 B,WML Yes Psychomotor
but not other No
Fox 199654 11 1 51.3 B ? No ?
Kaye 199755 30 3.6 84 B,H,TL Yes No ?
Mueller 199856 46 5 >65 B Yes ? ?
Jack 199858 24 1 81.04 H Yes ? ?
Fox 199959 15 6 55.3 B Yes ? ?
Schmidt 199960 273 3 60 WML Yes ? No
Ylikoski 200061 35 5 55-70 WML,TL Yes Yes No
Fox 200062 18 1 65 B Yes No ?
Jack 200064 58 3 80.4 H Yes Yes ?
Garde 200063 68 16 80.7 WML Yes Yes ?
Moffat 200065 26 2.7 68 H Yes ? ?

aWML = White matter lesions, B = Whole brain or other brain regions, H =

Hippocampal area, TL = Temporal lobe area.
bEvidence not provided.

individuals. Although volume loss and cognitive decline were present in the
normal groups, none of the studies found a positive correlation between MRI
volume changes and cognitive change.
This lack of association cannot be readily attributed to length of follow-
up since the measures were taken over long intervals (up to and including
five years). However, one explanation is that the age range of the subjects
was restricted (i.e. there was only a small range of ages in those examined),
thereby reducing the correlation. Similarly, it is possible that floor or ceiling
COGNITIVE CHANGES 89

Table 2. Longitudinal studies reporting CT changes and their associations with

cognitive change.

Study Number Length Age Brain Change Change Correlationb

of of (years) areaa in brain in
subjects study area cognitionb
(years)

Alzheimer’s Disease
Luxenberg 198770 18 5 62.8 V Yes ? Yes
De Leon 198968 50 3 71.2 VBR, V,
WML Yes ? ?
Burns 199167 63 1 79.3 V, CA Yes Yes Yes
Wippold 199166 60 4 72.6 V, CSF Yes ? No*
De Carli 199271 20 4.5 66 V Yes ? Yes
Jobst 199472 61 4 73.1 TL Yes Yes ?
Shear 199573 41 2.1 70.7 CSF Yes ? Yes
Meyer 200074 19 5.8 59.5 CA, V ? Yes ?

Cognitive decline, Mild Cognitive Impairment (MCI), or “at risk”

Bird 198669 5 4 >60 VBR,CA ? ? Yes
Meyer 200074 22 5.8 59.5 CA, V ? Yes ? Yes

Normal Individuals
Bird 198669 50 4 >60 VBR, V,
CA No No ?
Luxenberg 198770 12 5 65.1 V Yes ? ?
De Leon 198968 45 3 68.9 VBR, V,
WML Yes ? ?
Burns 199167
Wippold 199166 58 4 72.9 V,CSF No No ?
De Carli 199271 17 17 62 V Yes ? ?
Jobst 199472 47 1.7 62 TL Yes No ?
Shear 199573 35 2.6 67.4 CSF ? Yes No ?
Meyer 200074 224 5.8 59.5 CA, V Yes ? ?

aVBR= Ventricular brain ratio, V= Ventricular volume, CSF= Cerebrospinal volume,

TL= Temporal lobe, CA= Cerebral atrophy, WML= White matter lesions.
*Cognitive test here was the Clinical Dementia Rating Scale and this may not be a
valid measure of cognitive change.
bEvidence not provided.

effects in the tests result in a restricted range of test scores and this reduced
the size of the correlation. Since many of the studies used screening tests that
are known to have ceiling effects, this is a strong possibility. Another expla-
nation is that reliable tracing of brain structures is only possible once these
structures show significant involution or shrinkage. A major problem with
the reviewed studies is that the measures of both brain area and cognitive
90 THE AGEING BRAIN

domain are rather crude. The relationships between brain structural changes
and cognitive performance are best examined using tests that reflect specific
cognitive domains. This type of analysis has been undertaken in a recent
cross-sectional analysis. Raz et al.75 examined the relationship between brain
substrates and measures of working memory, explicit memory, priming and
executive functioning. Contrary to expectations, they found no relationship
between explicit memory and limbic brain volumes in a large sample of
113 individuals who ranged in age from 18–77 years. They suggested that
explicit memory might only be affected once extensive damage is incurred
in relevant brain structures, a finding consistent with the data in Table 1.
If this were the case, however, it leaves unanswered the question of which
brain structures or processes mediate the changes in cognitive performance
that are observed in “healthy” older adults. Loss of hippocampal volume
may not be responsible. Research examining brain volumes and cognitive
test scores in large community samples can be expected to clarify these
relationships, especially if these investigations include a range of specific
cognitive tests and examine limbic areas in addition to the hippocampus 75
and areas of the pre-frontal cortex.
Table 2 outlines the results from longitudinal studies using computed
tomography (CT) and measuring cognitive change. Although CT imaging is
less specific than MRI, the findings in Table 2 are highly consistent with the
MRI data. Table 2 demonstrates that there is loss of volume in brain areas as
assessed by CT in normal individuals, those with AD and those with mild cog-
nitive decline. As is expected, a change in cognition is less likely to be detected
in normal samples than those with dementia or mild cognitive impairment.
Almost all studies show an association between change in CT measures and
deterioration in cognitive performance for AD and MCD samples. The single
failure to find this association66 in the AD samples may well be due to the
insensitivity of the Clinical Dementia Rating Scale which was the “cognitive
test” used in the study. Consistent with the MRI findings shown in Table 1,
the association between changes in cognitive performance and changes in
brain volumes as measured by CT scans in non-clinical populations was not
established. However, for the CT findings as opposed to the MRI studies, this
is likely to be due to the lack of cognitive change observed in the individuals
in the normal groups who were followed up.
It was argued earlier that structural neuroimaging provides the opportu-
nity to examine relationships between cognitive and brain changes. In par-
ticular, this area of investigation may inform us about the causes of cognitive
ageing and the possible biological substrate of the “common factor”. At this
stage, these investigations have not provided clear evidence of the structures
which underlie memory or cognitive change in normal individuals. These
early longitudinal studies (Tables 1 and 2) have methodological limitations.
However, if these studies were indicative of the underlying associations, they
suggest that brain volume losses occur independently of cognitive change. In
particular, brain volume loss is apparent without corresponding cognitive
COGNITIVE CHANGES 91

change55,62,73 and where both cognitive and brain volume loss is observed
there is no correlation between the two.,53,61

Conclusions

Cognitive ageing is an exciting area of investigation. We know that on average

older people perform more poorly in areas of memory and speed than younger
people. There is evidence that the elderly are less reliable and less consistent
than younger people, that there is greater diversity of cognitive responses
in older age, and a number of important risk factors for decline have been
identified, in particular ApoE. Much is known about predictors of cognitive
change, but the new challenge is to identify the causes of cognitive deteriora-
tion. This process will be facilitated by longitudinal research. In recent years
there has been a renewed attempt to relate changes in biological processes,
indexed by measures of forced expiratory volume (FEV) and grip strength,
changes in brain structure as measured by white matter lesions, brain and
hippocampal volume and changes across cognitive domains. Further research
will involve the examination of a greater range of associations among changes
in cognitive and non-cognitive variables, including changes in indices of brain
function, cell metabolism, health and activities of daily living.

Acknowledgements

This chapter was supported by grant 973302 from the National Health and
Medical Research Council and by the Australian Rotary Health Research
Fund. Thanks are due to AF Jorm, AE Korten, AS Henderson, PA Jacomb
and B Rodgers for their contribution to the survey design, management and
analysis of the Canberra Longitudinal Study, and to S Hofer, Department of
Human Development and Family Studies,The Pennsylvania State University
and AJ Mackinnon, The Mental Health Institute of Victoria, and Department
of Psychological Medicine, Monash University, for their assistance with data
analysis.

References

1. Salthouse TA. Theoretical perspectives on Cognitive Aging. Hillsdale, NJ: Erl-

baum, 1991.
2. Schaie KW. Intellectual development in adulthood. The Seattle Longitudinal
Study. Cambridge, UK: Cambridge University Press, 1996.
3. Hultsch DF, Hertzog C, Dixon RA, Small, BJ. Memory change in the aged. Cam-
bridge, UK: Cambridge University Press, 1998.
4. Greenwood PM. The frontal aging hypothesis evaluated. J Int Neuropsych Soc.
2000; 6:705–726.
92 THE AGEING BRAIN

5. Flynn JR. The mean IQ of Americans: Massive gains 1932 to 1978. Psychological
Bulletin. 1984; 95:29–51.
6. Horn JL, Catell RB. Refinement and test of the theory of fluid and crystallized
intelligence. J Educ Psychol. 1966; 57:253–270.
7. Salthouse TA. Adult cognition. New York: Springer Verlag, 1982.
8. Wechsler DA. Manual for the Wechsler Adult Intelligence Scale Revised. New
York, The Psychological Corporation, 1981.
9. Deary IJ. Looking down on human intelligence. Oxford, UK: Oxford University
Press, 2000.
10. Baltes PB, Staudinger UM, Lindenberger U. Lifespan psychology: Theory and
Application to intellectual functioning. Annu Rev Psychol. 1999; 50:471–507.
11. Sattler JM. Age effects on Wechsler Adult Intelligence Scale-Revised tests. J Con-
sult Clin Psych. 1982; 50:785–6.
12. Christensen H, Mackinnon AJ, Korten AE, Jorm AF, Henderson AS, Jacomb P,
Rodgers B. An analysis of diversity in the cognitive performance of elderly com-
munity dwellers: Individual differences as a function of age. Psychol Aging. 1999;
14:365–379.
13. Sliwinski M, Buschke H. Cross-sectional and longitudinal relationships among
age, cognition, and processing speed. Psychol Aging. 1999; 14:18–33.
14. Anstey KJ, Luszcz MA, Sanchez L. A reevaluation of the common factor theory
of shared variance among age, sensory function, and cognitive function in older
adults. J Gerontol B Psychol Sci Soc Sci. 2001; 56:P3–11.
15. Korten AE, Jorm AF, Jiao Z, Letenneur L, Jacomb PA, Henderson AS, Chris-
tensen H, Rodgers B. Health, cognitive, and psychosocial factors as predictors of
mortality in an elderly community sample. J Epidemiol Commun H. 1999; 53:
83–8.
16. Small BJ, Backman L. Cognitive correlates of mortality: evidence from a popula-
tion-based sample of very old adults. Psychol Aging. 1997; 12:309–13.
17. Nelson, HE. National Adult Reading Test (NART). Berkshire, UK: NFER-Nel-
son, 1982.
18. Nelson EA, Dannefer D. Aged heterogeneity: fact or fiction? The fate of diversity
in gerontological research. Gerontologist. 1992; 32:17–23.
19. Morse CK. Does variability increase with age? An archival study of cognitive
measures. Psychol Aging. 1993; 8:156–64.
20. Rabbitt PA. Does it all go together when it goes? The nineteenth Bartlett Memo-
rial Lecture. Q J Exp Psychol. 1993; 46A:385–434.
21. Lindenberger U, Baltes P. Emergence of a powerful connection between sensory
and cognitive functions across the adult life span: a new window to the study of
cognitive aging? Psychol Aging. 1997; 12:12–21.
22. Kraemer HC, Kazdin AE, Offord DR, Kessler RC, Jensen PS, Kupfer DJ. Coming
to terms with the terms of risk. Arch Gen Psychiat. 1997; 54:337–43.
23. Anstey KJ, Christensen H. Education, activity, health, BP and APOE as predictors
of cognitive change in old age: A review. Gerontology. 2000; 46:163–177.
24. Albert MS, Jones K, Savage R, Berkman L, Seeman T, Blazer D., Rowe JW. Pre-
dictors of cognitive change in older persons: McArthur studies of successful aging.
Psychol Aging. 1995; 10:578–589.
25. Evans DA, Beckett LA, Albert MS, Herbert LE, Scherr PA, Funkenstein HH,
Taylor, JO. Level of Education and change in cognitive function in a community
population of older persons. Ann Epidemiol. 1993; 3:71–77.
26. Arbuckle TY, Maag U, Pushkar D, Chaikelson JS. Individual differences in trajec-
tory of intellectual development over 45 years of adulthood. Psychol Aging.1998;
13:663–675.
27. Prince M, Lewis G, Bird A, Blizard R, Mann A. A longitudinal study of factors
predicting change in cognitive test scores over time, in an old hypertensive popu-
lation. Psychol Med. 1996; 26:555–568
COGNITIVE CHANGES 93

28. Haan MN, Shemanski l, Jagust WJ, Manolio TA, Kuller L. The role of APOE
[epsilon]4 in modulating effects of other risk factors for cognitive decline in eld-
erly persons. J Am Med Assoc. 1999; 282:40.
29. Jonker C, Schmand B, Lindeboom J, Havekes LM, Launer LJ. Association
between apolipoprotein E epsilon4 and the rate of cognitive decline in com-
munity-dwelling elderly individuals with and without dementia. Arch Neurol
— Chicago, 1998; 55:1065–1069.
30. Feskens EJ, Havekes LM, Kalmijn S, de Knijff P, Launer LJ, Kromhout D. Apoli-
poprotein ε4 allele and cognitive decline in elderly men. Brit Med J, 1994; 309:
1202–1206.
31. Hyman BT, Gomez-Isla T, Briggs M, Chung H, Nichols S, Cohout F, Wallace
R. Apolipoprotein E and cognitive change in an elderly population. Ann Neurol,
1996; 40:55–66.
32. Yaffe K, Cauley J, Sands L, Browner W. Apolipoprotein E phenotype and cog-
nitive decline in a prospective study of elderly community women. Arch Neu-
rol—Chicago, 1997; 54:1110–1114.
33. Henderson AS, Easteal S, Jorm AF, Mackinnon AJ, Korten AE, Christensen H,
Croft L, Jacomb PA. Apolipoprotein E allele epsilon 4, dementia, and cognitive
decline in a population sample. Lancet, 1995; 346:1387–1390.
34. Small BJ, Basun H, Backman L. Three-year changes in cognitive performance as
a function of apolipoprotein E genotype: evidence from very old adults without
dementia. Psychol Aging, 1998; 13:80–87.
35. O’Hara R, Yesavage JA, Kraemer HC, Mauricio M, Friedman LF, Murphy GM.
The APOE epsilon4 allele is associated with decline on delayed recall performance
in community-dwelling older adults. J Am Geriatr Soc. 1998; 46:1493–1498.
36. Brayne C, Harrington CR, Wischik CM, Huppert FA, Chi LY, Xuereb JH,
O’Connor DW, Paykel ES. Apolipoprotein E Genotype in the prediction of cogni-
tive decline and dementia in a prospectively studied elderly population. Dementia.
1996; 7:169–74.
37. Helkala EL, Koivisto K, Hanninen T, Vanhanan M, Kervinen K, Kuusisto J,
Mykkanen L, Kesaniemi YA, Laakso M, Riekkinen P. Memory functions in
human subjects with different apolipoprotein E phenotypes during a 3-year
population-based follow-up study. Neurosci Lett. 1996; 204:177–180.
38. Carmelli D, Swan GE, LaRue A, Eslinger PJ. Correlates of change in cognitive
function in survivors from the Western Collaborative Group Study. Neuroepide-
miology. 1997; 16:285–295.
39. Dik MG, Jonker C, Bouter LM, Geerlings MI, van Kamp GJ, Deeg DJ. APOE-
epsilon4 is associated with memory decline in cognitively impaired elderly. Neu-
rology. 1999; 54:1492–1497.
40. Juva K, Verkkoniemi A, Viramo P, Polvikoski T, Kainulainen K, Kontula K,
Sulkava R. APOE epsilon4 does not predict mortality, cognitive decline, or
dementia in the oldest old. Neurology. 2000; 54:412–415.
41. Salthouse TA, Kausler DH, Saults JS. Age, self-assessed health status, and cogni-
tion. J Gerontol. 1990; 45:P156–160.
42. Nesselroade JR. The warp and the woof of the developmental fabric. In: Downs
R, Liben, L, Palermo DS, editors. Visions of aesthetics, the environment, and
development: The legacy of Joachim F. Wohlwill. New Jersey: Erlbaum, 1991;
213–240.
43. Anstey KJ. Sensorimotor and forced expiratory volume as correlates of speed,
accuracy, and variability in reaction time performance in late adulthood. Aging
Neuropsychol C. 1999; 6:84–95.
44. Li S-C, Aggen SH, Nesselroade JR, Baltes PB. Short term fluctuations in eldelry
people’s sensorimotor and memory performance: The MacArthur successful
ageing studies. Unpublished manuscript, Center for Lifespan Psychology, Max
Planck Institute for Human Development, Berlin, 2000.
94 THE AGEING BRAIN

45. Hultsch DF, MacDonald SWS, Hunter MA, Levy-Bencheton J, Strauss, E. Intrain-
dividual variability in cognitive performance in the elderly: Comparison of adults
with mild dementia, adults with arthritis, and health adults. Neuropsychology.
2000;14:588-598.
46. Hertzog C, Dixon RA, Hultsch, DF. Intraindividual change in text recall of the
elderly. Brain Lang. 1992; 42:248–269.
47. Salthouse TA, Hambrick DZ, McGuthry KE. Shared age-related influences on
cognitive and noncognitive variables. Psychol Aging. 1998; 13:486–500.
48. Lindenberger U, Baltes P. Sensory functioning and intelligence in old age: A strong
connection. Psychol Aging. 1994; 9:339–335.
49. Anstey KJ, Lord SR, Williams P. Strength in the lower limbs, visual contrast
sensitivity and simple reaction time predict cognition in older women. Psychol
Aging. 1997; 12:137–144.
50. Anstey KJ, Smith GA. Interrelationships among biological markers of aging,
health, activity, acculturation and cognitive performance in older adults. Psychol
Aging. 1999; 14:615–618.
51. Salthouse TA, Czaja SJ. Structural constraints on process explanations in cogni-
tive aging. Psychol Aging. 2000; 15;44–55.
52. Christensen H, Mackinnon AJ, Korten A, Jorm AF. The “common cause hypoth-
esis’ of cognitive aging: Evidence for a common factor but also specific associa-
tions of age with vision and grip strength in a cross-sectional analysis. Psychol
Aging. [In press].
53. Wahlund L, Almkvist O, Basun H, Julin P. MRI in successful aging: A 5-year
follow-up study from eighth to ninth decade of life. Magn Reson Imaging. 1996;
14:601–608.
54. Fox NC, Freeborough PA, Rossor MN. Visualization and quantification of rate
of atrophy in Alzheimer’s disease. Lancet. 1996; 348:94–97.
55. Kaye JA, Swihart T, Howieson D, Dame A, Moore MM, Karnos T, Camicioli
RM, Ball M, Oken B, Sexton G. Volume loss of the hippocampus and temporal
lobe in healthy elderly persons destined to develop dementia. Neurology. 1997;
48:1297–1304.
56. Mueller EA, Moore MM, Kerr DCR, Sexton G, Camicioli RM, Howieson DB,
Quinn JF, Kaye JA. Brain volume preserved in healthy elderly through the elev-
enth decade. Neurology. 1998; 51:1555–1562.
57. Fox NC, Warrington EK, Seiffer AL, Agnew SK, Rossor MN. Presymptomatic
cognitive deficits in individuals at risk of familial Alzheimer’s disease. A longitu-
dinal study. Brain. 1998; 121:1631–1639.
58. Jack CR, Petersen RC, Xu Y, O’Brien PC, Smith GE, Ivnik RJ, Tangalos EG, Kok-
men E. Rate of median temporal lobe atrophy in typical aging and Alzheimer’s
disease. Neurology. 1998; 51:993–999.
59. Fox NC, Scahill RI, Crum WR, Rossor MN. Correlation between rates of brain
atrophy and cognitive decline in AD. Neurology. 1999; 52:1687–1689.
60. Schmidt R, Fazekas F, Kapeller P, Schmidt H, Hartung H. MRI white matter
hyperintensities. Three year follow-up of the Austrian Stroke Prevention Study.
Neurology. 1999; 53:132–139
61. Ylikosky R, Salonen O, Mantyla R, Ylikosky A, Keskivaara P, Leskela M, Erkin-
juntti T. Hippocampal and temporal lobe atrophy and age-related decline in
memory. Acta Neurol Scand. 2000; 101:273–278.
62. Fox NC, Cousens S, Scahill R, Harvey RJ, Rossor MN. Using serial registered
brain magnetic resonance imaging to measure disease progression in Alzheim-
er’s disease. Power calculations and estimates of sample size to detect treatment
effects. Arch Neurol—Chicago. 2000; 57:339–344.
63. Garde E, Mortensen EL, Krabbe K, Rostrup E, Larsson HBW. Relation between
age-related decline in intelligence and cerebral white matter hyperintensities in
healthy octogenarians: a longitudinal study. Lancet. 2000; 356:628–34.
COGNITIVE CHANGES 95

64. Jack CR, Petersen RC, Xu Y, O’Brien PC, Smith GE, Ivnik RJ, Boeve BF, Tangalos
EG, Kokmen E. Rates of hippocampal atrophy correlate with change in clinical
status in aging and AD. Neurology. 2000; 55:484–489.
65. Moffat SD, Szekely CA, Zonderman AB, Kabani NJ, Resnick SM. Longitudinal
change in hipoocampal volume as a function of apolipoprotein E genotype. Neu-
rology. 2000; 55:134–136.
66. Wippold FJ, Gado MH, Morris JC, Duchek JM, Grant EA. Senile dementia and
healthy aging: a longitudinal study. Radiology. 1991; 179:215–219.
67. Burns A, Jacoby R, Levy R. Computed Tomography in Alzheimer’s disease: a
longitudinal study. Biol Psychiat. 1991;29:383–390.
68. De Leon MJ, George AE, Reisberg B, Ferris SH, Kluger A, Stylopoulos LA, Miller
JD, Regina ME, Chen C, Cohen J. Alzheimer’s disease: longitudinal CT Studies
of ventricular change. Am J Neuroradiol. 1989;10:371–376.
69. Bird JM, Levy R, Jacoby RJ. Computed tomography in the elderly: changes over
time in a normal population. Br J Psychiat. 1986;148:80–85.
70. Luxenberg JS, Haxby JV, Creasey H, Sundaram M, Rapoport SI. Rate of ven-
tricular enlargement in dementia of the Alzheimer type correlates with rate of
neuropsychological deterioration. Neurology. 1987; 37:1135–1140.
71. DeCarli C, Haxby JV, Gillette JA, Teichberg D, Rapoport SI, Schapiro MB. Lon-
gitudinal changes in lateral ventricular volume in patients with dementia of the
Alzheimer type. Neurology. 1992; 42: 2029–2036.
72. Jobst KA, Smith AD, Szatmari M, Esiri MM, Jaskowski A, Hindley N, McDonald
B, Molyneux AJ. Rapidly progressing atrophy of medial temporal lobe in Alzhe-
imer’s disease. Lancet. 1994; 343:829–830.
73. Shear PK, Sullivan EV, Mathalon DH, Lim KO, Davis LF, Yesavage JA, Tin-
klenberg JR, Pfefferbaum A. Longitudinal volumetric computed tomographic
analysis of regional brain changes in normal aging and Alzheimer’s disease. Arch
Neurol—Chicago. 1995; 52:392–402.
74. Meyer JS, Rauch G, Rauch RA, Haque A. Risk factors for cerebral perfusion,
mild cognitive impairment, and dementia. Neurobiol Aging. 2000; 21:161–169.
75. Raz N, Gunning-Dixon FM, Head D, Dupuis JH, Acker JD. Neuroanatomical
correlates of cognitive aging: evidence from structural magnetic resonance imag-
ing. Neuropsychology. 1998; 12:95–114.
76. Hofer, SM. Longitudinal studies of psychological aging. Department of Human
Development and Family Studies, The Pennsylvania State University [cited 2001
Aug 16] Available from: URL: http://www.personal.psu.edu/faculty/s/m/smh21/
index.htm.
Chapter 7

AGEING OF THE HUMAN

BRAIN AS STUDIED BY
FUNCTIONAL
NEUROIMAGING
Julian N. Trollor* and Perminder S. Sachdev

Introduction

A number of key molecular and structural changes occur in the brain as it

ages. The relationship between such changes and function of the aged brain
are poorly understood. Ageing brings with it preferential decline in fluid cog-
nitive abilities, with preservation of crystallised intellectual abilities1 (Chapter
6). In particular, age-related decline is seen in speed of information process-
ing, working memory and complex attention. The underlying mechanisms of
the decline in some cognitive domains remain obscure. Although structural
correlates of this phenomenon such as preferential decline in frontal lobe vol-
umes have been noted2 (Chapter 4), a less than robust relationship between
cognitive function and frontal lobe atrophy has been observed.3,4 Such dis-
crepancy emphasises the importance of exploring the functional correlates of
age-related neuropsychological decline.
Modern imaging techniques are appropriate tools to probe the inter-
relationship between molecular, structural and functional changes that
occur during the ageing process. One of the salient issues to be addressed
is that of resting cerebral metabolic and blood flow correlates of senescent
cognitive change. To this end, this chapter reviews relevant neuroimaging

*To whom correspondence should be addressed.

98 THE AGEING BRAIN

studies of ageing, as well as discussing the limitations of the literature.

Functional imaging correlates in special groups such as those with hyper-
tension, those at risk for cerebrovascular disease and those with known
risk factors for Alzheimer’s disease (AD) are explored. Not only are the
neuroimaging parameters at rest of importance, these techniques also enable
us to compare the effects of cognitive activation in young and old subjects.
Observations during activation assist in developing functional maps of nor-
mal and compromised brain function. The study of the effects of age on
the functional integrity of networks subserving certain cognitive functions,
especially memory, enables the integration of functional neuroanatomy and
cognitive theories of ageing. In addition to age-related decline in cognition,
the ageing process brings with it vulnerability to age-related neuropsychiat-
ric disorders. The application of functional imaging techniques provides an
opportunity to study key metabolic, blood flow and biochemical correlates
mediating this vulnerability.

Functional Imaging Techniques

SPECT & PET

Both positron emission tomography (PET) and single photon emission com-
puted tomography (SPECT) have been used to evaluate changes associated
with ageing and age-related neuropsychiatric disorders. These techniques
allow observation of a number of parameters related to cerebral function.
Regional cerebral blood flow (rCBF) can be measured with both techniques,
and cerebral metabolic rate of oxygen (rCMRO2) and cerebral metabolic rate
of glucose (rCMRglu) with PET.

Single photon emission computed tomography

SPECT has been used extensively to evaluate changes associated with age-
ing and age-related neuropsychiatric disorders. Contemporary brain SPECT
uses a rotating gamma camera to perform semiquantitative measurement of
rCBF by evaluating the relative distribution of a radiopharmaceutical agent
in the brain. SPECT detects single gamma rays (photons) and determines
the places of their origin based solely on their trajectories. SPECT is based
on the principle that some compounds will distribute in the body (in this
case the brain) in a way that reflects underlying physiologic function of a
particular region. Radiopharmaceuticals used to image the brain in SPECT
have differing characteristics, but most have relatively a long half life (T/2)
and are readily available commercially, a characteristic which makes SPECT
more widely available than PET. The most common of these in modern use
are 123I-IMP (iodine 123 N-isopropyl-p-iodoamphetamine), 99mTc-HMPAO
(technetium 99m hexamethylpropylamine oxime), 123I HIPDM (123I N,N
N-trimethyl-N-2-hydroxyl-3-methyl-5-iodobenzyl-1,3-propanediamine) and
133Xe (xenon 133).
CORRELATES OF BRAIN AGEING 99

Positron emission tomography

The fundamental principle of PET is the use of unstable isotopes that contain
an excess of protons in their nuclei and decay rapidly, emitting a positron,
which then travels a short distance before its annihilation by collision with
an electron. The collision results in the release of two gamma rays travelling
at 180o to each other. Coincident detection is registered when these gamma
rays strike a circular array of crystal detectors, from which the plane of the
annihilation event is determined. Information so gathered permits the math-
ematical conversion of data into quantitative measures reflecting the amount
of isotope present and its spatial distribution in tissue at a given time. As the
majority of radioisotopes used have a short T/2, on-site cyclotron production
is usually required. Isotopes used are typically biologically relevant, such as
15O (oxygen 15), 11C (carbon 11), 13N (nitrogen 13) and 18F (fluorine 18).

These isotopes can be used to label common biological matter (for example
H215O using 15O and fluorodeoxyglucose (FDG) using 18F), and can be used
to label drugs and receptor analogues.
The need of a cyclotron and a special camera makes PET an expensive
technology with only limited availability. 5 However, it has a number of
advantages over SPECT scanning. PET uses isotopes with short T/2 which
allow multiple scans in one individual, making it possible to perform activa-
tion studies using multiple variations of a task. While activation studies are
possible with SPECT, these are limited by the fact that more than two SPECT
scans (in a split-dose design) are not usually possible within a short period
because of the radiation exposure involved. PET offers superior spatial reso-
lution of the order of about 3–4 mm, while state-of-the-art SPECT does no
better than 7–8mm, but the latter is steadily improving with innovations in
technology. A limitation of SPECT is that quantitation of absolute blood-flow
values are not easily derived unless more invasive measures are undertaken,
e.g. sampling of carotid artery blood to assess absolute radio-activity tracer
counts. PET is also more versatile in its application, providing rCMRglu
(which SPECT does not), and being more readily adaptable to the study
of neurotransmitters and drugs. Because of their poorer resolution in com-
parison with MRI, specific regions on SPECT and PET images are generally
delineated after co-registering them with MRI scans, or by warping images
to fit a standard MRI template.

Functional magnetic resonance imaging (fMRI)

fMRI exploits the changes in microvascular oxygenation in response to
changes in blood flow to a region to estimate rCBF by the blood oxygen
level dependent (BOLD) technique. This is based on the inherent contrasting
magnetic properties of oxygenated versus deoxygenated haemoglobin. In its
oxygenated state, the iron in haemoglobin is diamagnetic, having a small
magnetic susceptibility effect. In the deoxygenated state, the iron in haemo-
globin has a large magnetic susceptibility effect and significantly disturbs
the surrounding magnetic field. At rest, a T2 gradient exists locally across
100 THE AGEING BRAIN

microvasculature, from a predominantly diamagnetic oxyhaemoglobin rich

environment to a paramagnetic deoxyhaemoglobin rich environment. In a
time-dependant manner following neuronal activation, there is an initial brief
decrease in oxygenated haemoglobin in the microvasculature, followed by a
more sustained increase. Thus, during brain activation (whether by motoric,
cognitive or pharmacological means), the relative excess of oxygenated hae-
moglobin produced by activation produces a T2 signal of greater intensity
compared with the resting or unactivated state. The measurement of the sig-
nal change represents the fundamental principle of BOLD fMRI.

Magnetic resonance spectroscopy (MRS)

Proton magnetic spectroscopy (1H-MRS) is a tool with a largely untapped
potential in the study of brain ageing. This non-invasive technique allows in
vivo estimations of specific brain chemical profiles, which can be compared
and contrasted across a variety of neuropsychiatric disorders. 1H-MRS can
detect N-acetyl containing compounds such as N-acetylaspartate (NAA),
which are putative markers of neuronal viability and density. Measurement
of choline containing compounds (Cho) (such as free choline, phosphocholine
and glycerophosphocholine) and creatine and phosphocreatine (Cr) allows
assessment of membrane synthesis and metabolism. Other metabolites such
as myoinositol (mI) are considered possible glial cell or intracellular toxin
markers. It is also possible to determine spectra for neurotransmitters such
as glutamate and GABA.

Studies of Cerebral Blood Flow and Metabolism in Healthy Ageing

Historical perspective
The history of modern techniques for measuring CBF can be traced to the early
work of Kety and Schmidt6, who first measured global CBF using an inhaled
nitrous oxide technique. This technique was invasive, requiring measurement
of arterio-venous difference in oxygen saturation, and allowed measurement
of average CBF per minute and derivation of oxygen consumption and cer-
ebrovascular resistance. In reviewing the series of sixteen early studies in which
this technique was used to measure CBF in ageing, it was concluded, “There
appears to be a rapid fall in over-all cerebral circulation about the time of
puberty, which continues through adolescence. From the third decade onward,
there is a more gradual decline in this function through middle and old age,”
and further that “cerebral oxygen consumption…shows similarly a rapid
and then a more gradual fall with advancing age”.7 The early studies were,
by today’s standards, seriously flawed methodologically. Nearly one-third
of these studies included hospitalised patients. Only four studies specifically
noted an absence of vascular disease and three studies noted absence of neu-
rological disease. In one study, some of the aged subjects were not considered
to be “mentally alert”.7 Although Kety himself concluded that the observed
CORRELATES OF BRAIN AGEING 101

changes may have in part been “the result of some age-dependant artifact”,7
these studies formed the basis for early conclusions that reduced cerebral blood
flow and oxygen consumption were a natural correlate of the ageing process.

Xenon studies of ageing

For the decade from the late 1970’s onward, the Xenon (133Xe) steady state
inhalation technique was commonly used to measure CBF. This technique
relied on detection of cerebral radioactivity using variable numbers of detec-
tors placed around the subject’s head after inhalation of 133Xe via a mask or
mouthpiece. Studies of ageing using 133Xe are summarised in Table 1. The
majority of these studies noted reduction in global CBF with age.8–13 An
exception was the negative study by Iwata and Harano,14 which may in part
be explained on the basis of the relatively small sample size and smaller age
range of the subjects. Despite their poor spatial resolution, regional variation
in CBF was also noted in some of these studies. A number of studies reported
anterior or middle cerebral artery territory predilection for this age-related
reduction in CBF.8,10,12,15 Others were not so specific, with one study report-
ing decline in rCBF in all regions except the brainstem,15 and another report-
ing preservation of rCBF only in the occipital regions.13 Only a single report
contradicts this general pattern of anterior preference on rCBF reduction.16
Laterality of effect was rarely explored, with one study demonstrating pre-
served Right – Left asymmetry with age, and another suggesting accentuation
of this affect in the frontal lobes with age.8
As can be seen from Table 1, a number of methodological issues are
immediately appreciable in relation to these 133Xe studies. Firstly, sample
selection was sub-optimal in several studies, with some subjects drawn from
clinical populations.8,9,11 Secondly, screening of individuals for inclusion was
performed with varying degrees of thoroughness, ranging from taking of a
medical history only in some of the earlier studies14–16 to thorough screen-
ing13 in more recent studies. Thirdly, factors known to influence CBF such
as visual or auditory stimulation and cognitive processing were rarely stand-
ardised between studies. Other factors such as the subject’s anxiety were not
acknowledged. Fourthly, only one of these early studies attempted to account
for the effects of cerebral atrophy on rCBF, noting a correlation between CBF
and an automated measure of atrophy derived from CT scan in subjects over
the age of 70 years. Finally, the spatial resolution of these earlier methods was
poor, making observed regional effects difficult to interpret. Taken together,
the comparability and generalisability of this collection of studies was poor.
However, there was some continued support of the notion of an age- depend-
ant decline in global CBF, with a suggestion of an anterior predominance of
this effect.

Technetium-HMPAO studies
The majority of recent SPECT studies of ageing have used 99mTc-HMPAO. It
is a highly lipophilic agent which crosses the blood-brain barrier and becomes
102 THE AGEING BRAIN

Table 1. Early Nitrous Oxide and Xenon Inhalation Studies of Cerebral Blood Flow in Ageing.

Study Sample N Age Tracer Eye/Ear Equipment/

Screening Procedure

Kety (1956) s 179 Mean age NO NS/NS NO technique

(Review)7 varied in 16
u studies from
5–93 years

Scheinberg s 32 2 groups: NO NS/NS NO technique

(1953)99 Ra: 18–36
u Ra: 38–79

Naritomi et s 60 3 Groups: Xe NS/NS 16 fixed

al. (1979)15 1. mean 28.5 detectors
u 2. mean 51.6 Res: NS
3. mean 56.6

Melamed et s 44 Ra: 19-79 Xe EO/EP 16 fixed

al. (1980)8 detectors
uu Res: NS

Iwata & ss 38 Ra: 23-65 Xe NS/NS Tomomatic 64

Harano Res: 17mm
(1986)14 u

Yamaguchi s 102 Ra: 26-81 Xe EC/EU Aloka RRG 526

et al. (1983)9 14 fixed
uu detectors
Res: NS

Matsuda sss 105 Ra: 19-80 Xe EC/NS 32 fixed detectors

et al. (1984)10 uuu Res: NS

Zemcov sss 44 Ra: 19-94 Xe EO/EU 32 fixed

et al, (1984)16 detectors
u Res: NS
 CORRELATES OF BRAIN AGEING 103

Method of Analysis Results

Kety & Schmidt Decline in CBFand oxygen consumption with age, rapidly
technique of arterial through adolescence then more gradual from third decade.
& venous sampling Kety considered that the changes may have been part of a
“the result of some age dependant artifact”

Kety & Schmidt CBF reduced with age.

technique of arterial No change in CMRO2
& venous sampling Arterial-venous O2 greater in aged
Cerebral vascular resistance greater in elderly
Middle aged v’s elderly comparison similar except CMRO2
greater in middle aged

Peak counts per 3 groups:

minute from 1. Young healthy group 28 subjects aged <40 years
individual detectors 2. Older healthy group of 18 subjects> 40 years
over lobar regions. 3. Risk factor group of 14 subjects over 40 years with 1 or more
risk for stroke:

Group 1 & 2: negative correlation of rCBF & age all regions

except brainstem cerebellum, most marked in MCA territory.
Group 3: same pattern as groups 1&2. No demonstrable effect
of vascular risk factors.

ISI used for rCBF Reduction in global and hemispherir CBF with age.
measure Reduction in rCBF in most regions, with anterior regions more
affected in left hemisphere.

ROI: method poorly Non-significant decrease in rCBF with age.

described. Large ROI’s
defined representing
cerebral vascular
territories.

ISI used for rCBF Global CBF reduced with age

measure. In old subjects 70+, correlation was seen between CBF and
Automated analysis of CT atrophy index. This was not observed for younger subjects.
scans allowed calculation
of craniocerebral index
(CCI) and brain volume
index (BVI)

ISI used for rCBF Global CBF globally decreased with age
measure. Greatest decrease for MCA territory

ISI used for rCBF Regional CBF decreased with age in bilateral temporal,
measure. parietal and occipital corticies.
Decrease in those with vascular risks more than others in some
regions, but differences failed to reach significance.

Table 1 continues
104 THE AGEING BRAIN

Table 1. Continued.

Study Sample N Age Tracer Eye/Ear Equipment/

Screening Procedure

Takeda ss 277 Ra: 19-88 Xe EC/EU Aloka RRG-526

et al. (1988)11 14 fixed
uuu detectors
Res: NS

Tsuda & sss 51 Ra: 19-77 Xe EC/EU 28 fixed

Hartmann detectors
(1989)12 uu Res: NS

Hagstadius sss 97 Ra: 19-68 Xe EC/EU NDS

& Risberg inhalation
(1989)13 uuu cerebrograph.
32 fixed
detectors
Res: NS

s, ss, sss = sample drawn from increasingly desirable source

u, uu, uuu = Increasingly detailed screening for optimal health
ISI: Initial Slope Index
NS: not specified
EO: eyes open; EC: eyes closed; EU: ears unplugged; EP: ears plugged
tracers: NO: nitrous oxide; Xe: Xenon inhalation; HMPAO: 99mTc-HMPAO
Res: Spatial Resolution at Full-width half-maximum (FWHM)
 CORRELATES OF BRAIN AGEING 105

Method of Analysis Results

ISI used for rCBF . Decrease in global CBF with age. Regional specificity
measure. not explored.
Automated analysis of CT Decrease in BVI noted in elderly, but not entered as covariate in
scans allowed calculation CBF analysis.
of craniocerebral index
index (CCI) and brain
volumeindex (BVI)

ISI used for rCBF Reduction in rCBF with age, with trend toward more reduction
measure in anterior regions. Reduction also observed in vascular CO
reactivity in aged subjects

Calculated grey matter Decreased global CBF with age.

flow and initial slope Regional decline in CBF in all areas except occipital region.
index (ISI). Hyperfrontality decreased with age.
Mean CBF higher in right hemisphere.
106 THE AGEING BRAIN

Table 2. Summary of Modern Resting SPECT Studies of Ageing.

Study Sample N Age Tracer Eye/Ear Equipment/

Screening Procedure

Waldemar ss 53 Ra: 21-83 HMPAO EC/EU Th45

et al. and Tomomatic 64
(1991)18 uuu Xe Res: NS

Markus sss 20 Ra: 21-76 HMPAO EO/EU GE/CGR

et al. Neurocam.
(1993)100 uu Res: NS

Swartz sss 47 Ra: 47-82 HMPAO EO/EU Headtome II

et al. for x and Res: NS
(1995)20 uu 27 for Xe
ceretec

Catafau sss 68 2 groups: HMPAO EO/ EU Elscint SP4-

et al. young mean HR.
(1996)19 uuu 29.5 Res: NS.
old mean
71.2

Mozley sss 44 Ra: 20-73 HMPAO EO/EU Picker 300s,

et al. (triple headed
(1997)21 uuu camera)
Res: NS

Krausz sss 27 Ra: 26-71 HMPAO EO/EU APEX ECT-

et al. 415, (single
(1998)17 headed)
u Res: 11mm

s, ss, sss = sample drawn from increasingly desirable source

u, uu, uuu = Increasingly detailed screening for optimal health
ISI: Initial Slope Index
NS: not specified
EO: eyes open; EC: eyes closed; EU: ears unplugged; EP: ears plugged
tracers: NO: nitrous oxide; Xe: Xenon inhalation; HMPAO: 99mTc-HMPAO
Res: Spatial Resolution at Full-width half-maximum (FWHM)
 CORRELATES OF BRAIN AGEING 107

Method of Analysis Results

Manual ROIs by Cortical atrophy but not age was a significant determinant of
reference to atlas. global CBF, accounting for 27% of variance.
CT scan ratings of Regional CBF in frontal & frontotemporal regions decreased
atrophy used for significantly with age.
analysis in 33 subjects

Positioning of square Increased assymetry of ROI uptake in elderly, reaching

ROI with reference to significance in temporal and temporoparietal regions only.
atlas. Uptake in both cortical and subcortical regions was less in
Normalised to injected elderly, but not significant. Non-significant reduction in frontal
dose and cerebellum. CBF.

Positioning of circular Xe: Global CBF stable with age. Decline in occipital CBF
ROI 1.9cm diameter. noted with age. In men only, global, frontal, temporal, occip
Tc-HMAPO regions & right hemishpere declines.
normalised to ROI Tc-HMPAO: no age related declines in global of regional CBF.
containing maximal
counts.

Manual ROIs drawn Highest relative CBF in cerebellum & occipital cortex.
directly onto slice. Effect of age on global CBF not reported.
Normalisation to both Decreased CBF in anterior regions, L frontal/temporal region
cerebellum and whole with age.
brain counts. Preservation of R>L assymetry with age.

Standardised ROI Regional CBF decreased with age in most cortical regions;
template. minimally so in ocipital and cerebellar ROI’s, maximally in
Counts normalised to frontal ROI’s.
whole brain. Regression analysis indicated non-linear model; broken stick
model with breakpoint median 36.6 yrs across regions.
Increase in white matter CBF noted.

Standardised ROI Subjects divided into young <50; old >50

template. Cerebellum normalisation: global decrease in rCBF with age.
Counts normalised to Regional decrease in left superior temporal, left basal ganglia,
cerebellum and whole and bilateral frontal, occipital and parietal ROI’s.
brain. Whole slice normalisation: regional decrease in occipital, right
frontal and increases in right thalamus, right anterior cingulate
and bilateral posterior cingulate.
Hemispheric perfusion difference R>L unaffected by age.
108 THE AGEING BRAIN

trapped intraneuronally by metabolic reduction, thus being distributed in

the brain in proportion to rCBF. One of the main advantages of 99mTc-
HMPAO over other SPECT radiotracers is that it remains intraneuronally
for at least six hours after intravenous administration, thus allowing images
to be acquired for some hours after the injection, without loss of resolution.
99mTc-HMPAO has favourable dosimetry, allowing improvement in image

quality at higher doses. An important development in SPECT technology has

been the availability of high-resolution dedicated neuro-SPECT scanners, sig-
nificantly narrowing the gap between SPECT and PET in achievable spatial
resolution.
Modern SPECT studies using 99mTc-HMPAO are listed in Table 2. These
studies contrast in some respects to the earlier Xenon studies. With the excep-
tion of the study by Krausz et al.,17 in which detail was omitted, assessment
of subjects has been sufficiently detailed to exclude confounding illnesses.
Improved spatial resolution has enabled more detailed region of interest
(ROI) analysis. Approaches to this analysis have involved either manually
tracing individualised regions onto subjects’ scans18,19 or overlaying of tem-
plates with predefined anatomical or geometric regions.17,20,21 The techno-
logical and methodological advances are reflected in results which challenge
the notion of age-dependent decline in global CBF, with a number of studies
reporting a negative finding.18,20
Where age effects on laterality of CBF have been explored, preservation of
R–L asymmetry has been reported.17,19 The most consistent regional effect of
age has been the demonstration of a reduction in frontal CBF.17-19,21 Decline
in temporal CBF18–20 has also been noted as an age effect. An interesting
finding, reported in one study, was that age-dependent decline in rCBF was
non-linear, with more reduction occurring up until middle age, and negligible
further reduction thereafter.21 If this were confirmed, there would be sig-
nificant implications for the age range included in future studies. A significant
limitation of these studies is that SPECT measures represent relative rather
than absolute CBF values. In obtaining this relative value, normalisation is fre-
quently obtained to either whole brain or cerebellum. Discrepancies in the way
in which the ageing process affects the cerebellum compared to other brain
regions may introduce some variability in findings, and this has been noted in
one study in which normalisation to both denominators was used.17

Resting 15O PET studies of ageing

The methodological details and results of key 15O-PET studies of normal
ageing are summarised in Table 3. PET investigations of cerebral correlates
of brain ageing share similar problems to that of SPECT. Earlier studies
featured samples drawn from hospital populations 22,23 or suffered from
poorly described or inadequate screening processes.22,24 Early studies were
performed with equipment offering poor spatial resolution.22–26
Despite recent improvement in a number of these methodological issues,
the heterogeneity of the methods of analysis makes comparison of studies
CORRELATES OF BRAIN AGEING 109

difficult, and limits the conclusions able to be drawn. The majority of studies
have used regions of interest (ROI) analysis, relying on poorly represented
anatomical definition on PET scans themselves to allow placement of stand-
ardised or geometric ROIs. Supprisingly, only a small number of studies27,28
have sought to co-register the MRI and PET images. In the study by Bentour-
kia et al.,27 however, the co-registration method (visual interpolation of the
two images) lacked the sophistication of other contemporary ROI analyses.
Only one 15O PET study has used a semiautomated statistical analysis pack-
age — Statistical Parametric Mapping (SPM; Wellcome Department of Behav-
ioural Neurology, Institute of Neurology, London) — for analysis. The issue
of correction of CBF and CMRO2 measurements for partial volume effects
has not been addressed in the majority of studies. Two studies have used
CT scan based methods to correct CBF and CMRO2 measures for the effect
of global atrophy.25,29 One study utilised an MRI-based method to correct
for partial volume effects for individual regions,28 offering a more rigorous
approach to partial volume correction.

Studies of global CBF and CMRO2

Although there are inconsistencies in the results from these studies with
respect to global changes in CMRO2, CBF and OEF with age, this is not
unexpected given the methodological disparities. A number of studies docu-
ment a global decline in CMRO2 with age.23,24,26,29–31 The magnitude of this
effect is relatively small, with various estimates of 0.3%23 0.37%,27 0.5%26
and 0.6% per annum.29 The failure of one study to find evidence of reduction
in CMRO2 with age deserves mention. Itoh et al25, showed linear decline with
age in a measure of cerebral atrophy (cerebrocranial index) but no decline in
either CBF or CMRO2. However, the age range of subjects was narrow, and
weighted toward middle age and older subjects between 50 and 85 years. In
view of the proposed non-linear decline in CBF with age proposed by Mozley
et al.,21 it is possible that the failure of this study to detect global reduction in
CBF or CMRO2 was related to the restricted age range of subjects included
in this study.
Modest decline in global CBF with age has been a common but less con-
sistent finding (see Table 3). In addition to the negative finding by Itoh et
al.25 with respect to both CBF and CMRO2, two studies demonstrated a
discrepancy between age effects on global CMRO2 and CBF.30,31 In these
two studies, CBF reduction was not seen with age, raising the possibility of
an uncoupling between CBF and CMRO2 as a factor of age.31 However, as
hypothesised by Yamaguchi et al.,30 such a discrepancy may be artifactual
and reflective of other age-associated changes known to affect CBF measure-
ment such as reduction in haematocrit and increase in PaCO2 with age. A
less consistent finding in 15O-PET studies is that of changes in oxygen extrac-
tion ratio (OER) with age. A number of studies have failed to find such a
relationship.22,29,30 In the study by Pantano et al.,23 a small but statistically
insignificant increase in grey matter OER of 7% was found between young
110 THE AGEING BRAIN

Table 3. Summary of Major Resting 15O PET Studies of Ageing.

Author Sample N Age Range (years) Eye/Ear Equipment/

Screening (sex ratio) Technique

Lebrun- s 19 19–76 EC/EU ECAT, measured

Grandie et (14M, 5F) attenuation.
al. (1983)22 u 15O Steady State
Inhalation
Res: n

Lenzi sss 27 Age not specified. NS/NS Equipment NS

et al. (NS) Divided into 2 groups: 15O Steady State
(1981)24 u Young < 50 (n=16) Inhalation
Old >50 (n=11) Res: NS

Pantano s 27 19–76 EC/EU ECAT II

et al. (19M, 8F) Divided into 2 groups: 15O Steady State
(1984)23 uu Young: av age 38 Inhalation
Old : av age 63 Res: n

Yamaguchi sss 22 26–64 EO/EU HEADTOME III

et al. (17M, 5F) Divided into 2 groups: 15O Steady State
(1986)30 u Young: av age 35.7 Inhalation
Old: av age 57.6 Res: n n n

Itoh sss 28 50–85 NS/NS ECAT II

et al. (17M, 11F) 15O Steady State
(1990)25 u Inhalation
Res: n

Leenders sss 30 22–82 EC/EO ECAT II

et al. (18M, 16F) 15O Steady State
(1990)26 uu Inhalation
Res: n

Burns & sss 14 51–85 NS/NS ECAT/931/08/12

Tyrrell (6M, 8F) 15O Steady State
(1992)101 uu Inhalation
Res: NS

Takada sss 32 27–67 EC/EU Equipment: NS

et al. (15M, 17F) 15O Steady State
(1992)31 uuu Inhalation
Res: NS

Martin sss 30 30–85 EC/EU ECAT/931/08/12

et al. (15M, 15F) 15O Steady State
(1991)32 uu Inhalation
Res: n n n
 CORRELATES OF BRAIN AGEING
Method of Analysis
Procedure
Circular ROI’s 10mm diamater
manually placed over CBF image
at point of highest CBF in lobe.
NS
Circular ROI’s traced on CBF
image and copied onto CMRO2
& OEF images
Normalised to mean of all ROI’s
Circular ROI’s placed with CT
superimposed for guidance
ROI measuring 3 x 7 pixels
CT brain used to calculate
cerebrocranial index (CCI)
Geometric ROI (rectangular for
cortical regions, circular for other
regions), individually placed.
Stereotactic co-ordinates based
on Talairach atlas used to obtain
CMRO2 values
ROI, method unstated
SPM
111
Results
Selected rCBF decrease with age temporosylvian,
medial frontal, medial occipital regions.
No correlation between regional CMRO2 or OER
and age.
rCBF & CMRO2 decline with age, most pronounced
in visual cortex & insula.
Increased OER with age.
rCBF and CMRO2 declined by 18% and 17%
respectively in grey matter. Largest decrease in frontal,
parieto-occipital and temporosylvian regions.
Non-significant increase in OEF in grey matter.
No difference in white matter between 2 groups.
Mean left hemisphere CMRO2 significantly lower in
older age group.
Significant correlation between CMRO2 and age
demonstrated.
No correlation between age and rCBF, or OEF.
CCI decreased linearly with age (Correlation –0.23)
CBF and CMRO2 unchanged with age and not
influenced by CCI
Decrease in CMRO2, CBF and CBV with age
(approx 0.5% per year), both in grey and white
matter.
Mostly non-significant increase in OER with age.
Decrease in CMRO2 in parietal lobe. In other
regions, this failed to reach significance.
Decrease in mean CMRO2 with age and in specific
ROI’s (bilateral putame, left supratemporal, left
infrafrontal and left parietal corticies).
Decreased CBF in left superior temporal cortex only.
Mean CBF unchanged with age.
RegionaL decrease in CBF in cingulate,
parahippocampal gyri, superior temporal gyri,
medial frontal gyri, parietal cortex bilaterally and in left
insular and inferior frontal gyrus.
Table 3 continues
112 THE AGEING BRAIN

Table 3. Continued.

Author Sample N Age Range (years) Eye/Ear Equipment/

Screening (sex ratio) Technique

Marchal sss 25 (NS) 20–68 EC/EU LETI TTV03

et al. 15O Steady State
(1992)29 uuu Inhalation
Res n n n

Eustache sss 25 20–68 EC/EU LETI TTV03

et al. (14M, 11F) 15O Steady State
(1995)102 uuu Inhalation
Res: n n n

Bentourkia sss 20 21–75 EC/EU ECAT EXACT-HR

et al. (13M, 7F) Divided into 2 groups: Subjects Both 15O H2O and
(2000)27 uuu Young: aged 21-36 asked to 18FDG injection
Old: aged 55–75 “avoid Res: n n n
focussing
their
mind on
anything”

Meltzer sss 27 19–76 EC/EU Siemens 951R/31

et al. (9M, 18F) 15O H O injection
2
(2000)28 uu Res: n n.

s, ss, sss sample drawn from increasingly optimal source (s= hospitalised subject;
ss outpatient clinic; sss volunteer)
u, uu, uuu increasingly rigorous screening (u history only, or unstated;
uu history & physical examination +/- laboratory tests;
uuu history, physical examination & CT/MRI brain or neuropsychological testing)
Res: Spatial Resolution at Full-width half-maximum (FWHM)
n = spatial resolution ≥ 15mm; n n = spatial resolution 8.6-14 mm;
n n n = spatial resolution ≤ 8.5mm
NS: not specified
EO: eyes open; EC: eyes closed; EU: ears unplugged; EP: ears plugged
 CORRELATES OF BRAIN AGEING
Method of Analysis
Procedure
ROI, circular, 14 pixel diameter
CT scans given atropy rating on
4 point scale
CT & PET images co-registered
(method undefined).
Circular ROI 14mm diameter
manually placed.
Normalisation of ROI values to
cerebellum.
MRI and PET images anatomically
matched by visual interpolation
ROI’s manually drawn on MRI
and adjusted for FDG study.
Transferred onto 15O study
MRI & PET coregistered using
using automated computerised
algorithm. ROI’s traced on MRI
& transferred to PET.
MRI based partial volume
correction method.
113
Results
Decrease in CMRO2 whole cortex and in multiple
cortical gyri (24/31) with age (approx –6% per decade).
Effect on whole cortex independent of cerebral
atrophy.
Decrease in CBF in 10/31 cortical gyri with age.
No change in OEF with age.
No decrease in CMRO2 or CBF in white matter or deep
grey matter structures.
Global decline in CMRO2 with age.
rCMRO2 values negatively correlated with age in
all neocortical regions and left thalamus.
Global decline in CBF with age (0.37% per year)
Decline in CBF was greatest in frontal regions and
least in Occipital cortex.
Preserved coupling of rCBF and rCMRGlu with age
Prior to partial volume correction: negative
correlation of mean CBF with age. Regional
statistically significant results in medial
orbitofrontal, lateral orbitofrontal, lateral
temporal regions.
After partial volume correction: mean CBF no longer
significantly correlated with age. Loss of regionally
significant results except for medial orbitofrontal
region.
114 THE AGEING BRAIN

subjects (mean age 38 years) and older subjects (mean age 63). A non-sig-
nificant increase of 0.35% per year in grey matter OEF was observed by
Leenders et al.26 The reported increase in OER by Lenzi et al.24 was modest
and not examined for statistical significance. In general, these results suggest
that OER may increase slightly with age, and may simply be a reflection of
reciprocal decreases in CBF.
Regional changes in CMRO2 and CBF with age have been examined in
a number of studies. An obvious discrepancy between white matter regions
and grey matter regions has emerged, with white matter CMRO2 and CBF
being largely unaffected by age.22,23,29,30 The most consistent regional effect
of age in cortical grey matter is that of decline in CBF and CMRO2 in selected
frontal and temporal regions.22,23,27,29,31,32 Reports of regional declines in
cortical CBF or CMRO2 have been predominantly bilateral, although some
studies have noted a left-sided31,32 or right-sided emphasis29 for particular
regions. An asymmetric left hemispheric decline in CMRO2 with age was
noted in one study.30

Resting FDG studies of ageing

Evaluation of the effect of ageing on global CMRglu has been undertaken in
a number of studies, with conflicting results (see Table 4). In an early study
in which the assessment methods for subject selection were not detailed,
Kuhl et al.,33 reported a decline in global CMRglu of 0.43% per year. A
number of studies have subsequently reported a global decrease in CMRglu
ranging from 0.21%34 to 0.6% per year.35 However, the reported decline in
CMRglu is uncertain. The screening of aged subjects appears to have been
inadequate in a number of studies in which global declines in CMRglu has
been described.33,34 Although most studies excluded those with established
hypertension, other cerebrovascular risk factors were often not assessed.
Several studies have used either CT scan36,37 or MRI scan27 as part of the
screening process, but this represents the exception rather than the rule. The
effect of cerebral atrophy on global CMRglu has been evaluated in a small
number of studies. Schlageter et al.37 employed an automated segmentation
technique to measure CSF volume on CT scan of the brains of their subjects.
Cerebral atrophy so measured was negatively correlated with CMRglu but
only explained 13% of the variance. No age effect on CMRglu was apparent
in this study. Another study initially revealed a decline in CMRglu with age,
which was no longer statistically significant after the effect of cerebral atro-
phy (as measured by semiquantitative ratings of MRI scans) was partialled
out38. In summary, there are methodologically sound studies demonstrating
both decline in CMRglu with age35,39 and no change with age.36,37 The vari-
ability of these findings indicates that perhaps a modest effect of age on global
CMRglu is being inconsistently determined as a result of methodological and
statistical differences between studies.
The comparison of FDG studies examining regional changes in CMRglu
with age is difficult owing to the different methods used to examine regional
CORRELATES OF BRAIN AGEING 115

effects. The majority of studies have employed ROI analysis. The spectrum of
ROI methods employed in these studies include: the poorly standardised pro-
cedure of tracing irregular regions directly onto the FDG study;40 the use of
the subject’s CT scan of the brain36,37 or an atlas41 as an anatomical guide for
ROI definition; the placing of standard geometric ROIs38,39,42 directly onto
the PET image and the tracing of regions onto individual’s MRI scan of the
brain and transferring onto the PET image.27 The recent use of standardised
packages such as SPM has enabled exploratory comparison across the whole
brain and enables easy comparison of results across studies using reference
to Talairach co-ordinates.
Despite the difference in methods of analysis for regional effects, only a
small number of studies have failed to find a significant regional effect of
age on CMRglu.36,40–42 A number of factors including poor resolution of
the older scanners and less sophisticated ROI analyses may explain many
of these negative findings. A distinct pattern of age related CMRglu decline
has emerged which is reminiscent of CBF and CMRO2 studies in ageing. The
most consistent pattern of reduction in CMRglu with age is that of frontal
reduction27,33,34,35,38,43–45 including that of anterior cingulate. 35,43 Other
regions showing reduction in CMRglu with age include specific anterior,
posterior and lateral temporal regions35,39,43,45 and parietal cortex.33,34,39,45
The effect of ageing on deep grey matter nuclei is occasionally reported as
being a decline, and in the study by Bentourkia et al., 27 the greatest CMR-
glu reduction was observed in the right striatum. The effect of ageing on
CMRglu in the cerebellum, occipital cortex and white matter appears to be
minimal.

Methodological Issues in Resting Studies

Defining healthy ageing

One of the key difficulties encountered in any study of the effects of ageing
is that of the confounding effects of pre-existing disease. Reducing the likeli-
hood that such confounding factors will influence results requires adoption
of a narrow definition of “normal ageing”, thereby excluding those with
risk factors for cerebral disease. Although for the sake of sample uniformity
conservative inclusion criteria may be desirable, there is a risk that findings
from any such study will reflect those of “elite ageing” rather than “normal
ageing”. Such results are less likely to be generalisable to the majority of the
aged population. Until more is known about the impact of common age-
related conditions on functional imaging findings, it would seem prudent
to select a conservative sample to minimise potential confounding effects of
systemic disorders on the results. The adequacy of the screening process prior
to enrolment of subjects can be questioned in many of the studies reviewed
in Tables 1–4. Appropriate clinician review including detailed history, physi-
cal examination, laboratory evaluation and neuropsychological assessment,
116 THE AGEING BRAIN

Table 4. Summary of Major Resting FDG Studies of Ageing.

Author Sample N Age Range (years) Eye/Ear Equipment/

Screening (sex ratio) Technique

Kuhl sss 40 18–78 EO/EU ECAT II

et al. (17M, 23F) Res: n
(1982)33 NS

De Leon NS 37 Divided into 2 EC/EU PET III

et al. (NS) groups: Res: n
(1984)42 NS Young: av age
26.1
Old: av age 66.6

Hawkins NS 8 18–68 NS/NS NeuroECAT

et al. (7M, 1F) Res: n n

Duara sss 21M 21–83 EC/EP ECAT II

et al Res: n
(1983)41 uu

Duara sss 40M 21–83 EC/EP ECAT II

et al. Res: n
(1984)36 uuu

Horwitz sss 30M Divided into 2 EC/EP ECAT II

et al. groups Res: n
(1986)103 NS Young: 20–32
Old: 64–83

Schlageter sss 49M 21–83 EC/EP ECAT II

et al. Res: n
(1987)37 uuu

Yoshii sss 76 21–84 EC/EU PETT V

et al. (39M, 37F) Res: n
(1988)38 uu

Hoffman sss 36 21–74 NS/NS NS

et al. (22M, 14F) Res: NS
(1988)45 uu
 CORRELATES OF BRAIN AGEING
Method of Analysis
Procedure
Not stated
Fudicial markers on PET and CT
brain used to match CT and PET
slices. Geometric ROI’s placed on
CT and transferred to PET image.
Irregular individual ROI’s
Method unspecified.
Individual ROI’s defined on PET
using atlas guide.
Individual ROI’s defined on PET
using atlas and patient’s CT as a
guide.
Individual ROI’s defined on PET
image with reference to brain atlas.
Partial correlation coefficiants
determined between each pair of
59 regions.
Individual ROI’s defined on PET
using individual’s CT as a guide.
CSF volume measured using
automated segmentation on
CT scan.
Geometric ROI’s placed
automatically over PET image and
manually adjusted to overlay
specific regions.
MRI scans of 58 subjects rated
semiquantiatively for atrophy.
ROI analysis, method unspecified.
117
Results
Global decline in rCMRGlu with age (0.43% per year).
Regional decline in rCMRGlu evidenced by reduction
in metabolic ration of superior frontal cortex to
parietal cortex.
No difference in global or regional rCMRGlu
between young and elderly subjects.
No change in rCMRGlu with age.
No significant change in global or regional
rCMRGlu for white or grey matter with age.
Study was an extension of above study
(Duara et al 1983).
Replicated above results.
No change in mean CMRGlu between groups.
Elderly subjects demonstrated a reduced number of
significant correlations between pairs of regions noted.
This reduction was especially noted in frontal-parietal
and parietal-parietal correlations.
CSF volume, ie cerebral atrophy was negatively
correlated with CMRGlu but accounted for no more
than 13% of variance.
No effect of age on CMRGlu, even after correction for
cerebral atrophy.
Divided into those with or without 1 or more risk
factors for thromboembolic stroke.
Significant lower mean CMRGlu in aged subjects
compared to young.
Age effect non-significant when effects of cerebral
volume and atrophy partialed out.
Atrophy accounted for 8.3% of variance of mean
CMRGlu.
CMRGlu not affected by presence of risk factors for
stroke.
RCMRGlu reduction with age in gyrus rectus, orbital
gyri, inferior frontal gyri, medial prefrontal cortex,
insula, superior parietal lobule & globus pallidus.
Table 4 continues
118 THE AGEING BRAIN

Table 4. Continued.

Salmon sss 25 Divided into 2 NS/NS NeuroEcat

et al. (NS) groups: Res: n n
(1991)104 uu Young: av age
25.8
Old: av age 60.1
DeSanti sss 72M Divided into 2 EO/EU Siemens CTI-931
et al. groups: Res: n n n
(1995)105 uu Young: av age
27.5
Old: av age 67.6

Moeller sss Group 1: Group 1: 21–90 Group 1: Group 1:

et al. 130 EC/EP Scanditronix PC
(1996)34 NS (62M, 68F) 1024-7B
Res: n n n
Group 2: Group 2: 24–77 Group 2: Group 2:
20 EO/EU Scanditronix
(10M, 10F) Superpett 3000
Res: n n n
Murphy sss 120 21–91 EC/EP Scanditronix PC
et al. (55M, 65F) 1024 7B
(1996)39 uu Res: n n n

Petit- sss 24 20–70 EC/EP LETI TTV03

Taboué (15M, 9F) Res: n n n
et al. (1998)35 uuu

Garraux sss 43 19–75 EC/EU Siemens CTI

et al. (25M, 18F) Divided into 2 951 R 16/31
(1999)43 uu groups: Res: n n n
Young: 19–28
Old: 47-75
Bentourkia sss 20 21–75 EC/EU ECAT EXACT-HR
et al. (13M, 7F) Divided into 2 Subjects Res: n n n
(2000)27 uuu groups: asked to
Young 21–36 “avoid
Old 55-75 focussing
their
mind on
anything”

s, ss, sss sample drawn from increasingly optimal source (s= hospitalised subject;
ss outpatient clinic; sss volunteer)
u, uu, uuu increasingly rigorous screening (u history only, or unstated;
uu history & physical examination +/- laboratory tests;
 CORRELATES OF BRAIN AGEING
ROI analysis, method unspecified.
Absolute values & ratios
normalised to mean cortical
values both examined.
Manually traced ROIs, some
anatomical & some geometric.
ROI method not described.
Used scaled Subprofle Model
(SSM) to examine regional
covariation with age.
Group 2:
Circular ROI template
superimposed on each patients
PET.
SPM analysis.
SPM analysis.
MRI and PET images
anatomically matched by visual
interpolation. ROI’s manually
drawn on MRI and adjusted for
FDG study. Transferred onto
15O study.
uuu history, physical examination & CT/MRI brain or neuropsychological testing). Res:
Spatial Resolution at Full-width half-maximum (FWHM) n = spatial resolution ≥ 15mm; n
n = spatial resolution 8.6-14 mm; n n n = spatial resolution ≤ 8.5mm
NS: not specified. EO: eyes open; EC: eyes closed; EU: ears unplugged; EP: ears plugged.
119
Global CMRGlu unchanged with age.
Absolute rCMRGlu values unchanged with age.
Normalised rCMRGlu reduced in frontal lobes and
increased in cerebellum with age.
Absolute rCMRGlu values reduced with age in frontal
and temporal lobes.
Dorsolateral frontal region demonstrated stronger
relationship with age than orbitofrontal region,
decreasing 2.6 and 2.2umoles/100g/min per decade
respectively.
Both groups, global CMRGlu decreased significantly
with age (0.21% per year). Regional declines in frontal
regions observed in group 1 only.
In group 1, relative decrease in frontal rCMRGlu was
associated with covariate relative increases in
parietooccipital association areas, basal ganglia,
brainstem and cerebellum.
Mean CMRGlu decreased with age.
Regional CMRGlu declined in frontal, temporal and
parietal ROI’s, with assymetry of this decline noted
(parietal L>R decline, frontal R>L decline).
L>R assymetry in frontal rCMRGlu was significantly
less in women compared with men.
Global decline in rCMRGlu with age (6% per decade).
Regional decline in rCMRGlu in most areas except
occipital cortex and right cerebellum. Most marked age
related decline seen bilaterally in perisylvian
temporoparietal and anterior temporal regions, insula,
inferior and postero-lateral frontal region, anterior
cingulate, head of caudate, anterior thalamus.
Frontal CMRGlu decreased in elderly subjects in
bilateral dorsolateral prefrontal and medial prefrontal
areas including anterior cingulate, left lateral premotor
area, Broca’s area and left insular, right superior temporal
gyrus.
Global decline in rCMRGlc (0.34% per year).
Decline in rCMRGlc greatest in right striatum and
least in cerebellum.
Preserved coupling of rCBF and rCMRGlc with age.
120 THE AGEING BRAIN

is required to minimise the chance of inclusion of subjects with disorders

affecting cerebral function.

The resting state

The majority of SPECT and PET studies of ageing rely on evaluation of “rest-
ing” rCBF or rCMRglu. The notion of what represents an adequate “resting
state” has varied significantly across studies. The common measures of ensur-
ing reduced sensory input by plugging ears and patching eyes are not always
adopted. Limited attempt has been made to reduce the effects of cognitive
processing. Mental processing occurring during scanning may vary substan-
tially between subjects and is likely to be influenced by factors such as anxiety
provoked by the procedure, mood at time of scanning, etc. This uncontrolled
mental activity may in turn significantly influence rCBF or rCMRglu. An
approach to potentially overcome this confound and minimise inter-subject
variability is the introduction of a standardised cognitive task.

Accounting for cerebral atrophy

The comparison of ageing and youthful brains brings with it the challenge
of taking account of the effects of structural changes of ageing on functional
data. The most important issue is the effect of brain atrophy. As the resolu-
tion of both PET and SPECT fall significantly short of structural imaging
techniques such as MRI, functional data will be influenced by the effect of
atrophy due to partial-volume averaging. If such an effect is not taken into
account, functional data can considerably underestimate rCBF and rCMR-
glu in the elderly, resulting in misinterpretation of results. This effect may
account for some of the positive findings regarding reductions in rCBF and
rGMRgl in aged compared to young subjects. There have been several differ-
ent methods used to attempt to correct for this source of error. Simple tech-
niques include correction using an atrophy score derived from visual ratings
of structural scans, correction by using global measures of atrophy such as
VBR and more complicated correction by using a MRI-based segmentation
of images and generation of regional correction factors.
Meltzer et al.28 attempted to address this issue with surprising results.
Their study used manually traced regions of interest on MRI that were trans-
ferred to PET images using an automated image registration algorithm. An
MRI-based partial volume correction coefficient was applied to the CBF data.
This correction coefficient was derived for individual regions after segmenta-
tion of the MRI into brain and cerebrospinal fluid, creation of a binary data
set and smoothing of MRI data to approximate resolution of the PET scan.
After corrections were applied, most age-associated changes in cerebral blood
flow failed to reach significance. The exception was that CBF continued to
demonstrate a statistically significant negative correlation with age in the
mesial orbitofrontal region. Such a finding suggests that previous studies may
have overestimated age-related reduction in CBF and CMRO2. However, as
prefrontal atrophy is about twice that found in the temporal or parietal neo-
CORRELATES OF BRAIN AGEING 121

cortex44 atrophy effects are unlikely to account for all age effects on CBF or
CMRO2 noted in previous studies, uncorrected for atrophy or partial volume
effects. These findings challenge the notion of a large age-specific decrease in
brain blood flow and metabolism, and underscore the importance of correc-
tion for possible confounding factors.

Summary of resting SPECT and PET studies

Methodological discrepancies between studies limit the conclusions that can
be drawn from this body of literature. However, a general pattern is appreci-
able which both SPECT and PET studies of CBF and PET studies of CMRglu
share. With respect to global effects of the ageing process, a modest effect is
demonstrable in some studies. Regional effects appear to mirror known pat-
terns of age-related pathological change, including regional effects of atro-
phy. Although atrophy may partially account for regional CBF, CMRO2 and
CMRglu effects observed with age, it does not appear to be the sole arbiter
of these changes. Regional effects demonstrated by resting functional imaging
studies add support to theories of cognitive ageing which espouse decline in
frontal lobe function as a mediator of age-related change.

Resting PET and SPECT Studies in “At Risk” Groups

The major focus so far has been on defining the changes associated with
“healthy ageing”. The rigorous screening of subjects and exclusion of those
with risk factors for conditions such as cerebrovascular disease means that
such studies may be poorly representative of the general population. A limited
but expanding literature is exploring influence of age-associated phenomena
such as cerebrovascular disease and mild cognitive impairment on functional
imaging parameters.

Vascular risk factors

A few studies of healthy ageing have evaluated the effects of cerebrovascular
risk factors on CBF. Measuring with the nitrous oxide technique, earlier stud-
ies46 failed to demonstrate an effect of hypertension alone on CBF or CMRO2.
However, in the presence of systemic arteriosclerosis or frank cerebrovascular
disease, both parameters declined and cerebrovascular resistance increased.
Using 133Xe inhalation SPECT, Naritomi et al.15 divided their elderly group
into those with and those without risk factors for stroke. Although an age-
related effect was observed on rCBF, the presence of cerebrovascular risk
factors did not alter rCBF. In an FDG PET study of ageing, Yoshii et al.38
divided their sample of 76 subjects into those with and without risk factors
for thrombo-embolic stroke. No appreciable effect on CMRglu was noted in
the presence of stroke risk factors. In a study of 60 aged individuals ranging
from 65 to 84 years, Claus et al.47 evaluated the effect of cerebrovascular
risk factors, quantified indicators of atherosclerosis and cerebral atrophy on
122 THE AGEING BRAIN

CBF as measured by 99mTc-HMPAO. An age-related decline in rCBF in tem-

poral and parietal cortex was observed with age, and this relationship was
preserved after adjustment for the effect of cerebrovascular risk factors. The
relationship between age and rCBF disappeared after correction for fibrino-
gen levels and measures of carotid atherosclerosis. The authors proposed that
age-related declines in CBF might be related to atherosclerosis rather than to
vascular risk factors per se.

Hypertension
133Xe inhalation was used in a 36-month prospective study evaluating the

effect of antihypertensive treatment on CBF in 12 individuals with mild

hypertension at baseline.48 As a group, an overall increase in CBF was
demonstrable at 6, 12 and 24 months after initiation of antihypertensive
treatment. This difference was no longer appreciable by 36 months. Four
subjects developed overt signs of cerebrovascular disease over the course of
follow-up. The CBF values of these four individuals showed decline from
the 24-month evaluation onward, whereas the asymptomatic group contin-
ued to demonstrate improved CBF values throughout the study period. In
a cross-sectional study, Nobili et al.49 showed that hypertensives who were
neurologically asymptomatic, especially when untreated, had focal or diffuse
cerebral hypometabolism. More recently, utilising FDG PET, Salerno et al.50
compared a group of 17 elderly hypertensives compared with 25 age-matched
non-hypertensive controls. In the hypertensive group, a significant reduction
in FGD uptake was demonstrable in regions supplied by basal ganglia perfo-
rating arteries and at the middle cerebral/anterior cerebral artery watershed.
Although data are limited, it appears that the presence of hypertension alone
has a modest but appreciable influence on CBF and CMRglu. This effect may
be regional, affecting areas most vulnerable to the effects of ischaemia such
as long perforating vessels and “watershed” regions.

White matter hyperintensities (WMHs)

Hyperintensities on T2-weighted MRI brain imaging are common in the white
matter of elderly individuals. The functional significance of these has been a
matter of dispute and extensive investigation. In a study of 51 healthy indi-
viduals between 19 and 91 years, De Carli et al.51 found that when WMHs
comprised >0.5% of intracranial volume, they were associated with cognitive
deficits and reduced frontal lobe blood flow on PET scanning. In another
study52 it was shown that cortical metabolic dysfunction was related to
ischemic subcortical lesions, both lacunar infarcts and non-infarction WMHs,
in patients with vascular dementia. Metabolism in the frontal cortex may be
particularly dependent on pathologic alterations of subcortical nuclei. In a
more recent study of 231 individuals53 without overt neurological disease, the
most striking relationship was that observed between periventricular white
matter hyperintensities (PVHs) measured by semiquantitative analysis and
CMRglu measured by region of interest analysis. CMRglu values showed
CORRELATES OF BRAIN AGEING 123

a progressive reduction with increasing grades of PVHs. The relationship

between CMRglu and severity of PVHs was significant for multiple cortical
and subcortical regions. A less strong relationship was observed between
CMRglu and deep white matter or basal ganglia hyperintensities. The rela-
tionships between MRI hyperintensities, cerebrovascular risk factors and
functional imaging measures awaits further exploration, but promises to be
important in understanding the determinants of age-related decline in CBF
and CMRglu.

Groups at risk of dementia

Functional imaging techniques have been used to evaluate individuals who
are at risk of age-related disease processes such as AD. The most informative
findings have come from those studies in which asymptomatic subjects at
risk of AD are compared with healthy age-matched controls. In a study of 24
asymptomatic first degree relatives from familial AD pedigrees, Kennedy et
al.54 demonstrated deficits in global and regional CMRglu compared with age
matched control subjects not at risk of AD. The regional pattern of CMRglu
abnormality was similar to, but less severe than that of affected familial AD
controls. In a PET study of cognitively normal subjects with family history of
AD,55 the effect of apolipoprotein A (ApoE ε4) was determined by compar-
ing 22 ApoE ε4-negative individuals with 11 subjects homozygous for ApoE
ε4. Subjects in both groups performed equally well on neuropsychological
measures. Global CMRglu was not different between groups. Those sub-
jects who were homozygous for ApoE ε4 demonstrated reduced rCMRglu in
posterior cingulate, parietal, temporal and prefrontal regions compared to
ApoE ε4-negative controls. A further study from the same group56 evaluated
10 cognitively normal individuals who were heterozygous for ApoE ε4 and
15 ApoE ε4-negative individuals with a family history of AD. Subjects were
followed longitudinally and underwent FDG PET at baseline and two-year
follow-up. Cognitive deterioration was not observed over the study period.
However, greater CMRglu declines were observed in the ApoE ε4 hetero-
zygous subjects in temporal, posterior cingulate and prefrontal cortex as well
as parahippocampal gyrus and thalamus, in comparison to ApoE ε4-nega-
tive subjects. Taken together, these preliminary studies of ApoE ε4 positive
individuals suggest that functional imaging changes are seen which parallel
those of AD rather than normal ageing. These results underscore the poten-
tial of ApoE status as a confounding factor in studies of normal ageing. The
screening of aged individuals for ApoE status prior to inclusion in functional
imaging studies of healthy ageing is therefore highly recommended. This topic
is described in greater detailed in Chapter 15.

Cognitive impairment
Those with mild cognitive impairment (MCI) may be at risk of developing
AD, with estimates of conversion varying from 10% to 15% per year.57 The
degree to which functional imaging may assist in the prediction of those at
124 THE AGEING BRAIN

risk of conversion, has yet to be properly explored. In preliminary reports,

significant temporo-parietal abnormalities have been noted with SPECT in
those with MCI.58,59 In a PET study, Small et al.60 evaluated 12 cognitively
normal subjects with age associated memory impairment (AAMI) and posi-
tive family history of AD, who were heterozygous for ApoE ε4 genotype and
compared them to a similar group of 19 ApoE ε4-negative individuals. CMR-
glu was significantly lower in both right and left parietal regions in the ApoE
ε4-positive group. In addition, left-right parietal asymmetry was significantly
higher in those at risk of AD and who were ApoE ε4-positive, than those
without ApoE ε4. These results suggest that functional imaging may be of
future utility in identification of subjects at risk of later cognitive decline.
However, at present, the positive predictive value of SPECT and PET in those
with MCI or AAMI is questionable.59

Activation Studies in Ageing

Introduction
In healthy young subjects, motor, cognitive and pharmacological challenges
have been shown to result in discrete changes in rCBF (using SPECT), rCMR-
glu or rCMRO2 (using PET) and BOLD signal using fMRI. These techniques
are now being applied to ageing and age-associated dysfunction, with a par-
ticular focus on cognitive activation. A range of cognitive and non-cognitive
tasks has proved suitable for probing functional integrity in aged subjects.
Activation procedures have included motoric,61 photic,62 working memory,63
visual attention,64 frontal/executive,65 word identification,66 spatial orienta-
tion,67 visual encoding and retrieval68 and verbal encoding and retrieval69
tasks. Intuitively, many of these tasks have been chosen on the assumption
that some age-dependent decline occurs within that particular cognitive
domain. There are several aims of this exercise. Firstly, there is a possibil-
ity that the activation process may unmask changes not appreciable at rest.
Secondly, there is a desire to study the functional neuroanatomy of the age-
ing brain, and to examine functional neuroimaging correlates of age-related
neuropsychological changes. Thirdly, the studies aim to identify individuals
at increased risk of age-related neuropsychiatric diseases.

Potential advantages of activation tasks

One advantage of introducing a cognitive task during scanning is the reduc-
tion in intra-subject variability of functional data. As mental activity and
sensory stimulation have cerebral metabolic and blood flow correlates, stand-
ardisation of sensory, motor and cognitive activity via an activation proce-
dure will minimise these potential confounds. The influence of these variables
over reliability of cerebral metabolic measurements has been demonstrated by
Duara et al.70 who re-scanned nine normal subjects at rest and seven normal
subjects during an activation task. Resting scans were performed with eyes
CORRELATES OF BRAIN AGEING 125

closed and covered and ears open. Activation scans involved viewing a series
of coloured pictures projected onto a screen, and depressing a foot pedal
with the right or left foot, depending on whether the subject did, or did not,
like a picture. Within subject variability for normal subjects for the repeated
scans performed at rest was marked, with correlation coefficient for mean
rCMRglu of 0.001, indicating virtually no correlation. However, in repeated
activated scans in normal subjects, there was a correlation coefficient of
0.703. Metabolic ratios for specific regions also varied considerably between
resting scans, but less so between activated scans. A study by Skolnick et al. 71
used 133Xe inhalation to measure rCBF in elderly normal subjects at rest and
during verbal and spatial tasks during two scans, separated by an average of
nine weeks. Global CBF was reduced in the repeated baseline scans, but this
reduction was not evident in the repeated activated scans. Regional CBF was
consistent between the two activated scans. These results suggest that meas-
urement of CBF and CMRglu is reproducible and reliable during activated
states and may be less so at rest.

Motor and sensory stimulation

In a simple reaction time task with BOLD fMRI, D’Esposito et al.72 compared
32 young subjects (mean age 22.9 years) and 20 elderly subjects (mean age
71.3 years). They found a reduced number of suprathreshold voxels and
lower signal-to-noise ratio (SNR) in the elderly, but no significant group dif-
ferences in the shape of the haemodynamic response. The authors suggested
that an age effect on the haemodynamic coupling between neural activity and
BOLD signal change may need to be taken into account in interpretation of
age effects demonstrated by functional imaging techniques. Calautti et al.61
used 15O PET to investigate age effects in a cued thumb-to-index finger oppo-
sition task. An over-activation in the superior frontal cortex in aged relative
to young subjects was noted, which, in the authors’ opinion, could not be
fully explained by differences in resting CBF in this area between young and
old subjects.
In a BOLD contrast fMRI experiment, Ross et al.62 examined the response
to photic stimulation in a group of 9 healthy elderly subjects (mean age 71
years) and healthy young subjects (mean age 24 years). The mean BOLD signal
response to photic stimulation in the visual cortex was significantly reduced
in aged subjects. There was a trend for elderly subjects, without atrophy, to
demonstrate less profound reduction than those with atrophy. During presen-
tation of checkerboard stimuli to 11 young subjects (mean age 23 years) and
11 elderly subjects (mean age 66 years), Huettel et al.73 noted an earlier peak
in the haemodynamic response, smaller spatial extent of activation and lower
SNR in the elderly subjects. The amplitude, form and refractory properties
of the haemodynamic response were similar across groups, and the authors
concluded that the smaller spatial extent of activation was attributable to
the lower SNR in the elderly subjects. In a study of 12 young and 14 older
subjects, Levine et al.74 examined CBF correlates during passive viewing of
126 THE AGEING BRAIN

black and white formed and unformed textures. Subjects were scanned in
two states: during viewing of random (formed and unformed) textures or
formed textures. Random versus formed texture viewing was associated with
activation outside the ventral occipitotemporal pathway (predominantly in
left anterior cingulate, medial, middle and superior frontal gyri) in the older
subjects. Comparison with the younger group demonstrated reduced activa-
tion in the older subjects of left precuneus, middle temporal and posterior
cingulate and of the right parahippocampal gyrus. Taken together, these
results suggest an age-related change in processing of visual stimuli, which
may represent a decrement in the efficiency of visual processing with age.
Grady et al. have examined age-related changes in object and spatial visual
processing in a series of 15O PET experiments. An initial study contrasted a
face-matching task with a spatial location-matching task.75 Results from this
study suggested that functional separation of dorsal (occipitoparietal) versus
ventral (occipitotemporal) pathways subserving spatial relations and object
discrimination respectively was apparent in both aged and young subjects.
However, functional separation of these two systems appeared less distinct
in aged individuals. A further study replicated these findings,76 and in addi-
tion demonstrated greater activation by older adults during face matching
in bilateral dorsolateral prefrontal cortex, fusiform gyrus, inferior frontal
gyrus and left insula as well as left middle temporal gyrus. During location
matching, older subjects again activated a more widespread network in bilat-
eral prefrontal cortex, bilateral fusiform gyri as well as left occipitotemporal
cortex and inferior parietal cortex. Such changes were taken to represent
reduced processing efficiency of prestriate occipital cortex with age and
hence increased utilisation of additional networks in order to compensate
this reduction in efficiency.
Similar results were obtained in a 15O PET study evaluating the effects of
age on selective and divided visual attention by Madden et al.64 During the
divided attention task only, subjects activated a network of regions including
occipitotemporal, occipitoparietal and prefrontal regions. Activation patterns
in younger subjects were relatively greater in the occipitotemporal pathway
and for the older subjects greater in prefrontal regions. A further study by
Grady et al.77 evaluated the effect of age on a task of degraded and non-
degraded face perception. Results similar to the earlier study76 were obtained
for the non-degraded face matching. Analysis of the activation patterns from
the degraded face-matching task revealed that different regions were posi-
tively correlated with task performance in the old compared with the young
subjects. In the old subjects, activity in the posterior occipital cortex, thala-
mus and hippocampus showed positive correlation with task performance.
In the younger subjects this correlation occurred in the fusiform gyrus, sug-
gesting that brain networks subserving success in this task differed between
young and old subjects.
Using a short-term visual memory task in which subjects discriminated
pairs of vertical sinusoidal gratings of differing spatial frequency, McIntosh
CORRELATES OF BRAIN AGEING 127

et al.78 determined age-related differences in activated networks. Older and

younger subjects performed equally well on the task. Although a common
pattern of activation was seen in many regions (occipital, temporal and infe-
rior prefrontal cortex and caudate), older subjects were observed to activate
additional distinct regions (medial temporal and dorsolateral prefrontal cor-
tices). Activity in these additional regions was related to task performance in
the older subjects, suggesting a role for these additionally activated networks
in maintenance of performance.

Working memory
In the first report of ageing effects on visual working memory, Grady et
al.63 used 15O PET to evaluate rCBF during a delayed match to sample task.
Independent of age, a common pattern of activation was noted with delay,
including increased rCBF in left anterior prefrontal and decreased rCBF in
the ventral extrastriate cortex. Less activation was seen in the right ventro-
lateral prefrontal cortex, and greater activation was seen in left dorsolateral
prefrontal cortex in the aged group. A subsequent 15O PET study utilising
two verbal working memory tasks79 demonstrated similar networks of activa-
tion in older and younger subjects. However, younger subjects showed more
right dorsolateral prefrontal activation during one task and older subjects
demonstrated greater left dorsolateral prefrontal activation during another
working memory task. Taken together, these two studies support the notion
that increased activation observed in the older subjects is a reflection of com-
pensatory strategies required to overcome cognitive inefficiency in working
memory occurring with age.

Complex tasks
Age effects have been examined during more complex tasks such as card sort-
ing. An 15O PET study by Nagahama et al.80 utilised a modified card sorting
task. This study revealed age effects of reduced ability to activate left dorso-
lateral prefrontal cortex, left inferior parietal lobule, left striate and prestriate
cortex, bilateral precuneus, left occipital cortex and left cerebellum. In aged
subjects a negative correlation between perseverative errors and activation
was noted in several of these regions including left dorsolateral prefrontal
cortex. A subsequent 15O PET utilising the Wisconsin Card Sort Test (WCST)
and Raven’s Progressive Matrices (RPM) for activation65 has demonstrated
age specific reductions in activated networks. Age-related reduction in acti-
vation was seen in the dorsolateral prefrontal cortex with WCST and in the
inferolateral temporal cortex with RPM. In addition, aged subjects activated
areas that were normally suppressed in younger individuals. These areas
included right parahippocampal gyrus with WCST and polar and medial
portions of the prefrontal cortex in both WCST and RPM. Reduced activa-
tion in key areas normally subserving these more complex neuropsychological
functions appears linked with decrements on task performance. However, it
remains unclear whether enhanced activation in regions normally suppressed
128 THE AGEING BRAIN

by younger individuals represents use of alternative cognitive strategies or

inefficiency of operating networks subserving these cognitive tasks.65
Madden et al.66 investigated the effect of age on a visual word identifica-
tion task using 15O PET in 10 young and 10 older subjects. Activity repre-
senting passive encoding of letter strings was observed in left frontal, striate
and inferior temporal cortex, with more prominent activations observed in
frontal and temporal regions in younger subjects. Several decreases in rCBF
were also observed, with these being greater in the older subjects in the right
superior frontal region and greater in the younger subjects for the right ante-
rior thalamus, right posterior cingulate and insula. The activity associated
with the semantic retrieval component of the task (distinguishing word from
pronounceable non-word) also revealed an age effect. More prominent left
occipital activation was observed in younger subjects, and a more prominent
deactivation was observed in left superior frontal, anterior cingulate and
lateral aspect of inferior temporal gyrus.

Memory tasks

Encoding
Correlates of age-related declines in encoding ability for visual material has
been examined in a study by Grady et al.68 using 15O PET in 10 young sub-
jects (mean age 25 and 10 older subjects (mean age 69). In this study, older
individuals activated the left ventral temporal cortex with encoding of faces
but failed to activate the network seen in younger individuals (anterior cin-
gulate, left prefrontal cortex, left temporal cortex and right medial temporal
lobe including hippocampus). The poor activation in older subjects was
attributed to failure of the encoding process. In a more complex task, encod-
ing of face/name pairs was examined in a small 15O PET study.81 Activation
during encoding was seen in bilateral occipital association areas, extending
into parietal lobes bilaterally. No activation was seen in the hippocampus
on either side. Reduced rCBF was observed in right temporal, frontal and
anterior cingulate regions and in the left superior temporal gyrus. No age
effect was identified, although this may have been attributable to the small
sample size.
Studies examining encoding of verbal material have also shown some
age-effects. During the encoding phase of a word pair task, Cabeza et al.69
demonstrated that younger subjects had higher activation in the left prefron-
tal and occipito-temporal regions than older subjects. Somewhat in contrast
to this study are the activation effects noted by Madden et al.82 During an
encoding task involving a living/non-living judgement of nouns, no significant
activation was observed in the younger subjects relative to a baseline task.
However, in the aged group, activation was observed in bilateral prefrontal
cortex, left thalamus, fusiform gyrus and parahippocampal gyrus. Direct con-
trast between young and older subjects revealed a significant difference for
thalamus only. Reductions during the encoding task were not seen in young
CORRELATES OF BRAIN AGEING 129

subjects but were seen in several regions in the elderly group, including left
anterior cingulate and right inferior parietal lobule. Direct comparison of the
two age groups failed to reveal significant differences in deactivation patterns.
The demonstration of age-related contrasts in patterns of activation during
encoding in this study stands in some contrast to the studies of Grady et al.68
and Cabeza et al.,69 and raises the possibility that age related differences in
rCBF during encoding may vary across tasks rather than as a function of age
per se.82

Retrieval
A number of studies have examined the effects of age on functional imaging
correlates of recall of verbal material. However, the comparability of results
is limited by the widely differing paradigms employed. Cabeza et al.69 used
a word pair task to study recognition and cued recall with 15O PET. Effects
of these “retrieval” tasks revealed that whilst younger subjects primarily
demonstrated a unilateral (right) sided effect for retrieval of verbal material,
older subjects demonstrated a more bilateral pattern of frontal activation.
Retrieval was examined in the noun recognition experiment undertaken by
Madden et al.82 Young subjects appeared to activate a network including
right prefrontal cortex, left middle frontal gyrus and left thalamus, whereas
older subjects activated bilateral prefrontal cortex, inferior parietal lobule,
left inferior temporal cortex and left cerebellum. Direct comparisons revealed
statistically significant increase in activation for younger subjects in the tha-
lamus only, and in prefrontal regions in the elderly. Deactivation was also
examined and, although deactivated regions were different in young and older
subjects, no statistically significant differences were apparent. Using a word
stem completion paradigm, Backman et al.83 explored 15O PET correlates
of a word stem completion task, incorporating baseline, priming and recall
components. Activity attributed to the recall component of the task in both
young and elderly included bilateral increases in prefrontal cortex and ante-
rior cingulate. Somewhat unexpectedly, older subjects activated perirhinal
cortex bilaterally. The two explanations posited for this medial temporal
lobe activity were that either optimal cued recall in elderly subjects involved
use of strategic search strategies not utilised by younger subjects or that older
subjects were continuing to encode and consolidate the information even
during the cued recall task. The medial temporal lobe activation nevertheless
stands in some contrast to the majority of studies of recall in both aged and
young subjects.
In a large 18FDG study of 70 subjects, Hazlett et al.84 examined cerebral
metabolic activity during performance of a serial verbal learning task, with-
out reference to a resting condition. Good performance was associated with
higher metabolic activity in the frontal lobes for in the younger subjects and in
the occipital lobes for the older subjects. An age related decrement in cerebral
metabolism in the frontal lobes was observed, which remained significant
after correction for cerebral atrophy. The shift in metabolic activity away
130 THE AGEING BRAIN

from anterior patterns of activation in the elderly was taken to indicate real-
location of networks invoked by the task. Comparisons of these results with
15O PET studies are made difficult by the combination of encoding and recall

components in this experiment.

Finally, Cabeza et al.85 examined age effects on activation during a verbal
retrieval task, evaluating both ability to recognise a previously studied word
from a distractor (content recognition), and ability to determine which of two
words had been presented more recently in the study list (recall of temporal
order). For content recognition, activation was observed in ventromedial tem-
poral regions in both groups, suggesting a limited effect of age on strategic
retrieval processes. However, age effects included reduced activation of right
prefrontal regions and enhanced activation of left prefrontal cortex. This
effect was thought to represent a compensatory process enacting semantic
processing in the elderly. An alternative explanation, that of increasing left
prefrontal activity with increasing demands of the task, was not supported
by performance data (which indicated equal performance across age groups)
or by comparison of hard versus easy blocks included in the experiment. For
the temporal order task, an increase was seen in the right prefrontal cortex
of young, but not older subjects, an effect that may reflect general age effects
on frontal function or regionally specific effects of age on temporal-order
retrieval.

Summary of activation studies

A number of studies have evaluated the effect of age on cerebral activation.
There is evidence, from studies employing simple motor and sensory para-
digms, of an age-related decline in cerebral activation, an effect that may have
its basis in dysfunction in pathways subserving sensory and motor functions.
For cognitive processes, the effect of ageing on patterns of activation is partly
dependent on the task under consideration. A unifying observation is that
during many cognitive tasks, elderly subjects activate brain networks that
are similar to those activated by young subjects. However, the extent of this
activation is reduced in elderly subjects. With some exceptions, this statement
can be made of tasks involving visual processing, short-term visual memory,
visual and verbal working memory, verbal recall and complex cognitive tasks
such as card sorting. Reduced ability to activate certain brain regions has
also been noted as a feature of ageing, and is often seen in association with
reduced performance of the task under consideration. Activation of a number
of regions additional to those seen in younger subjects has also been observed.
In encoding and recall tasks, a more bilateral pattern of prefrontal activation
has generally been observed.
Alterations in the pattern of prefrontal activation are of interest, given the
propensity of these regions to structural change with age. Enhanced activation
in prefrontal regions may be an attempt to compensate for reduced functional-
ity of this brain region with age, but this is a simple explanation that requires
more study. In particular, the relationship between prefrontal activation and
CORRELATES OF BRAIN AGEING 131

success during task performance requires further evaluation. Activation of

additional networks outside the prefrontal cortex has also been observed. At
times, this extra-frontal activation has been associated with maintenance of
performance. This suggests that functional activation of additional regions
may reflect engagement of different cognitive strategies, perhaps by way of
compensation for age related inefficiencies in processing.
In addition to age differences in activation, an age effect on deactivation
of various regions has been observed, with older subjects showing less strong
deactivation in usually deactivated networks, along with additional deacti-
vation in other areas. Deactivation may be a way in which optimisation of
cognitive performance occurs in health, thus suggesting inability to streamline
performance with age. A more detailed analysis of deactivation with respect
to task performance will allow the potential implication of this age-related
change in deactivation pattern to be elucidated.

Magnetic Resonance Spectroscopy (MRS)

Much of the work related to the effect of ageing on MRS-defined brain

metabolites has focused on the neurodevelopmental period 86,87 or young
adulthood.87 The studies of the elderly are limited by: small numbers, cross-
sectional design, limited sampling of brain tissue, different methods and
discrepant findings that warrant further studies before definitive conclusions
can be made.
The relative concentration of N-acetylaspartate (NAA), which is the major
marker of neurones, has been the focus of some investigations. Results have
been variable, with reports of a reduction in NAA with age in the basal
ganglia88 and the grey and white matter89–91 in some, but not all92,93 earlier
studies. More recent studies94–96 have reported no reduction in NAA in the
elderly, although this is still an inconsistent finding.97 These discrepancies
could be related to methodological differences and the age range studied.
Since MRS does not yield absolute values, metabolites are quantified in refer-
ence to an internal standard. The commonly used standard is total creatine
(Cr), but this is not invariant with age, thereby reducing the value of the
NAA/Cr ratio as a measure of age-related change. The other reference used
is MRI-visible water content, which accounts for >95% of tissue water in the
brain.98 While one study reported no change in the water content of the brain
with ageing,98 another study reported a significant reduction in brain water
with ageing.94
Total creatine (Cr), a marker of the energetics of neurones and glia, has
been reported to decrease with age in the basal ganglia88 and increase with age
in the frontal white matter,93 frontal gray matter,94 parietal white matter96
and both grey and white,95 but some studies have reported no change.93,96
Choline-containing compounds (Cho) have been shown to increase with
age in some studies,89,94,95 decrease according to others88,93,97 and remain
132 THE AGEING BRAIN

unchanged in some studies.90,96

The heterogeneity of the above findings suggests that this field is still in its
infancy and much work remains to be done. The potential of MRS has there-
fore not been fully exploited in this field. MRS studies in the healthy elderly
are important if a normative database is to be developed for MRS to be used
as a diagnostic investigation for neuropsychiatric disorders in the elderly, in
particular dementia.

Conclusions

The increasing sophistication of functional imaging studies, particularly those

involving activation techniques, is extending our understanding of how the
brain is affected by the ageing process. Although many methodological pitfalls
exist, this body of literature has enabled an appreciation of more subtle age
effects than were initially thought to exist. At rest, age results in small, region-
ally specific declines in CBF and CMRglu, maximal in the frontal regions,
which mirrors structural and neuropsychological alterations observed with
advancing years. The differing patterns of prefrontal activation observed with
age across a variety of cognitive activation tasks further highlights a key role
for the frontal lobes in mediation of the age related cognitive changes seen in
health. Specific but contrasting changes in functional integrity are beginning
to be demonstrated in association with risk factors for age-related diseases
such as AD and cerebrovascular disease. There are many areas in which func-
tional imaging techniques may contribute to the further unravelling of secrets
of the ageing brain. A potential role exists for these technologies in assessing
the effects of interventions designed to modify or delay the ageing process.
Early identification of those at risk of age-related neuropsychiatric disease
might also be a realistic future role for these tools.

References
1. Baltes P, Staudinger U, Lindenberger U. Lifespan psychology: Theory and applica-
tion to intellectual functioning. Annu Rev Psych. 1999; 50:471–507.
2. Raz N, Gunning-Dixon F, Head D, Dupuis J, McQuain J, Briggs SD, Loken WJ,
Thornton AE, Acker JD. Selective aging of human cerebral cortex observed in
vivo: Differential vulnerability of the prefrontal gray matter. Cereb Cortex. 1997;
7:268–282.
3. Raz N, Gunning-Dixon F, Head D, Dupuis J, Acker J. Neuroanatomical correlates
of cognitive aging: Evidence from structural MRI. Neuropsychology. 1998; 12:
95–114.
4. O’Donnell K, Rapp P, Hof P. Preservation of prefrontal cortical volume in behav-
iorally characterized aged macaque monkeys. Exp Neurol. 1999; 160:300–310.
5. Schuckit MA. An introduction and overview to clinical applications of neu-
roSPECT in psychiatry. J Clin Psychiat. 1992; 53 Suppl:3-6.
6. Kety SS, Schmidt CF. Nitrous oxide method for the quantitative determination of
cerebral blood flow in man: Theory, procedure and normal values. J Clin Invest.
1948; 27:476–483.
CORRELATES OF BRAIN AGEING 133

7. Kety SS. Human cerebral blood flow and oxygen consumption as related to aging.
J Chron Dis. 1956; 3(5):478–486.
8. Melamed E, Lavy S, Bentin S, Cooper G, Rinot Y. Reduction in regional cerebral
blood flow during normal aging in man. Stroke. 1980; 11:31–35.
9. Yamaguchi T, Hatazawa J, Kubota K, Abe Y, Fujiwara T, Matsuzawa T. Cor-
relations between regional cerebral blood flow and age-related brain atrophy:
a quantitative study with computed tomography and the xenon-133 inhalation
method. J Amer Geriatr Soc. 1983; 31:412–416.
10. Matsuda H, Maeda T, Yamada M, Gui LX, Tonami N, Hisada K. Age-matched
normal values and topographic maps for regional cerebral blood flow measure-
ments by Xe-133 inhalation. Stroke. 1984; 15(2):336–342.
11. Takeda S, Matsuzawa T, Matsui H. Age-related changes in regional cerebral
blood flow and brain volume in healthy subjects. J Amer Geriatr Soc. 1988; 36(4):
293–297.
12. Tsuda Y, Hartmann A. Changes in hyperfrontality of cerebral blood flow and
carbon dioxide reactivity with age. Stroke. 1989; 20:1667–1673.
13. Hagstadius S, Risberg J. Regional cerebral blood flow characteristics and varia-
tions with age in resting normal subjects. Brain Cognition. 1989; 10:28–43.
14. Iwata K, Harano H. Regional cerebral blood flow changes in aging. Acta Radiol
— Suppl. 1986; 369:440–443.
15. Naritomi H, Meyer JS, Sakai F, Yamaguchi F, Shaw T. Effects of advancing age
on regional cerebral blood flow. Studies in normal subjects and subjects with risk
factors for atherothrombotic stroke. Arch Neurol. 1979; 36:410–416.
16. Zemcov A, Barclay L, Blass JP. Regional decline of cerebral blood flow with age
in cognitively intact subjects. Neurobiol Aging. 1984; 5:1–6.
17. Krausz Y, Bonne O, Gorfine M, Karger H, Lerer B, Chisin R. Age-related changes
in brain perfusion of normal subjects detected by 99mTc-HMPAO SPECT. Neu-
roradiology. 1998; 40:428–434.
18. Waldemar G, Hasselbalch SG, Andersen AR, Delecluse F, Petersen P, Johnsen
A, Paulson OB. 99mTc-d,l-HMPAO and SPECT of the brain in normal aging. J
Cerebr Blood F Me. 1991; 11:508–521.
19. Catafau AM, Lomena FJ, Pavia J, Parellada E, Bernardo M, Setoain J, Tolosa E.
Regional cerebral blood flow pattern in normal young and aged volunteers: A
99mTc-HMPAO SPET study. Eur J Nucl Med. 1996; 23:1329–1337.
20. Swartz JR, Lesser IM, Boone KB, Miller BL, Mena I. Cerebral blood flow changes
in normal aging — SPECT Measurements. Int J Geriatr Psych. 1995; 10:437–
446.
21. Mozley PD, Sadek AM, Alavi A, Gur RC, Muenz LR, Bunow BJ, Kim HJ, Stecker
MH, Jolles P, Newberg A. Effects of aging on the cerebral distribution of tech-
netium-99m hexamethylpropylene amine oxime in healthy humans. Eur J Nucl
Med. 1997; 24:754–761.
22. Lebrun-Grandie P, Baron JC, Soussaline F, Loch’h C, Sastre J, Bousser MG. Cou-
pling between regional blood flow and oxygen utilization in the normal human
brain. A study with positron tomography and oxygen 15. Arch Neurol. 1983; 40:
230–236.
23. Pantano P, Baron JC, Lebrun-Grandie P, Duquesnoy N, Bousser MG, Comar D.
Regional cerebral blood flow and oxygen consumption in human aging. Stroke.
1984; 15:635–641.
24. Lenzi GL, Frackowiak RS, Jones T, Heather JD, Lammertsma AA, Rhodes CG,
Pozzilli C. CMRO2 and CBF by the oxygen-15 inhalation technique. Results
in normal volunteers and cerebrovascular patients. Eur Neurol. 1981; 20:285–
290.
25. Itoh M, Hatazawa J, Miyazawa H, Matsui H, Meguro K, Yanai K, Kubota K,
Watanuki S, Ido T, Matsuzawa T. Stability of cerebral blood flow and oxygen
metabolism during normal aging. Gerontology. 1990; 36:43–48.
134 THE AGEING BRAIN

26. Leenders KL, Perani D, Lammertsma AA, Heather JD, Buckingham P, Healy MJ,
Gibbs JM, Wise RJ, Hatazawa J, Herold S. Cerebral blood flow, blood volume
and oxygen utilization. Normal values and effect of age. Brain. 1990; 113 (Pt 1):
27–47.
27. Bentourkia M, Bol A, Ivanoiu A, Labar D, Sibomana M, Coppens A, Michel
C, Cosnard G, De Volder AG. Comparison of regional cerebral blood flow and
glucose metabolism in the normal brain: effect of aging. J Neurol Sci. 2000; 181:
19–28.
28. Meltzer CC, Cantwell MN, Greer PJ, Ben Eliezer D, Smith G, Frank G, Kaye WH,
Houck PR, Price JC. Does cerebral blood flow decline in healthy aging? A PET
study with partial-volume correction. J Nucl Med. 2000; 41):1842–1848.
29. Marchal G, Rioux P, Petit-Taboue MC, Sette G, Travere JM, Le Poec C, Courthe-
oux P, Derlon JM, Baron JC. Regional cerebral oxygen consumption, blood flow,
and blood volume in healthy human aging. Arch Neurol. 1992; 49:1013–1020.
30. Yamaguchi T, Kanno I, Uemura K, Shishido F, Inugami A, Ogawa T, Murakami
M, Suzuki K. Reduction in regional cerebral metabolic rate of oxygen during
human aging. Stroke. 1986; 17:1220–1228.
31. Takada H, Nagata K, Hirata Y, Satoh Y, Watahiki Y, Sugawara J, Yokoyama E,
Kondoh Y, Sishido F, Inugami A. Age-related decline of cerebral oxygen metabo-
lism in normal population detected with positron emission tomography. Neurol
Res. 1992; 14 (2 Suppl):128–131.
32. Martin AJ, Friston KJ, Colebatch JG, Frackowiak RS. Decreases in regional cer-
ebral blood flow with normal aging. J Cerebr Blood F Met. 1991; 11:684–689.
33. Kuhl DE, Metter EJ, Riege WH, Phelps ME. Effects of human aging on patterns
of local cerebral glucose utilization determined by the [18F]fluorodeoxyglucose
method. J Cerebr Blood F Met. 1982; 2:163–171.
34. Moeller JR, Ishikawa T, Dhawan V, Spetsieris P, Mandel F, Alexander GE, Grady
C, Pietrini P, Eidelberg D. The metabolic topography of normal aging. J Cerebr
Blood F Met. 1996; 16:385–398.
35. Petit-Taboue MC, Landeau B, Desson J, Desranges B, Baron J. Effects of healthy
aging on the regional cerebral metabolic rate of glucose assessed with statistical
parametric mapping. Neuroimage. 1998; 7:176–184.
36. Duara R, Grady C, Haxby J, Ingvar D, Sokoloff L, Margolin RA, Manning RG,
Cutler NR, Rapoport SI. Human brain glucose utilization and cognitive function
in relation to age. Ann Neurol. 1984; 16(6):703–713.
37. Schlageter NL, Horwitz B, Creasey H, Carson R, Duara R, Berg GW, Rapaport
SI. Relation of measured brain glucose utilisation and cerebral atrophy in man. J
Neurol Neurosur Ps. 1987; 50:779–785.
38. Yoshii F, Barker WW, Chang JY, Loewenstein D, Apicella A, Smith D, Boothe
T, Ginsberg MD, Pascal S, Duara R. Sensitivity of cerebral glucose metabolism
to age, gender, brain volume, brain atrophy, and cerebrovascular risk factors. J
Cerebr Blood F Met. 1988; 8:654–661.
39. Murphy DG, DeCarli C, McIntosh AR, Daly E, Mentis MJ, Pietrini P, Szczepanik
J, Schapiro MB, Grady CL, Horwitz B, Rapoport SI. Sex differences in human
brain morphometry and metabolism: an in vivo quantitative magnetic resonance
imaging and positron emission tomography study on the effect of aging. Arch Gen
Psychiat. 1996; 53(7):585–594.
40. Hawkins RA, Mazziotta JC, Phelps ME, Huang SC, Kuhl DE, Carson RE, Metter
EJ, Riege WH. Cerebral glucose metabolism as a function of age in man: influence
of the rate constants in the fluorodeoxyglucose method. J Cerebr Blood F Met.
1983; 3:250–253.
41. Duara R, Margolin RA, Robertson-Tchabo EA, London ED, Schwartz M, Ren-
frew JW, Koziarz BJ, Sundaram M, Grady C, Moore AM. Cerebral glucose uti-
lization, as measured with positron emission tomography in 21 resting healthy
men between the ages of 21 and 83 years. Brain. 1983; 106:761–775.
CORRELATES OF BRAIN AGEING 135

42. de Leon MJ, George AE, Ferris SH, Christman DR, Fowler JS, Gentes CI, Brodie
J, Reisberg B, Wolf AP. Positron emission tomography and computed tomography
assessments of the aging human brain. J Comput Assist Tomo. 1984; 8:88–94.
43. Garraux G, Salmon E, Degueldre C, Lemaire C, Laureys S, Franck G. Compari-
son of impaired subcortico-frontal metabolic networks in normal aging, subcor-
tico-frontal dementia, and cortical frontal demential. Neuroimage. 1999; 10:
149–162.
44. Murphy D, DeCarli C, McIntosh A, Daly E, Mentis M, Pietrini P, et al. Age-
related differences in volumes of subcortical nuclei, brain matter, and cerebro-
spinal fluid in healthy men as measured with magnetic resonance imaging (MRI).
Arch Gen Psychiat. 1996; 53:585–594.
45. Hoffman JM, Guze BH, Hawk TC. Cerebral glucose metabolism in normal
individuals: effects of aging, sex and handedness. Neurology. 1988;38 (suppl 1):
167.
46. Shenkin HA, Novak P, Goluboff B, Soffe AM, Bortin L. The effects of aging,
arteriosclerosis, and hypertension upon the cerebral circulation. J Clin Invest.
1953; 32:459–465.
47. Claus JJ, Breteler MMB, Hasan D, Krenning EP, Bots ML, Grobbee DE, Van
Swieten JC, Van Harskamp F, Hoffman A. Regional cerebral blood flow and
cerebrovascular risk factors in the elderly population. Neurobiol Aging. 1998;
19:57–64.
48. Meyer JS, Rogers RL, Mortel KF. Prospective analysis of long term control of
mild hypertension on cerebral blood flow. Stroke. 1985; 16:985–990.
49. Nobili F, Rodriguez G, Marenco S, De Carli F, Gambaro M, Castello C, Pon-
tremoli R, Rosadini G. Regional cerebral blood flow in chronic hypertension: a
correlative study. Stroke. 1993; 24:1148–1152.
50. Salerno JA, Mentis MD, Grady CJ, Rapoport SI, Schapiro MB. Positron emis-
sion tomographic studies of brain function in older men with chronic essential
hypertension. J Am Geriatr Soc. 2001; 40.
51. De Carli C, Murphy DGM, Tranh M, Grady CL, Haxby JV, Gillette JA, Salerno
JA, Gonzales-Aviles A, Horwitz B, Rapaport SI. The effect of white matter
hyperintensity volume on brain structure, cognitive performance, and cerebral
metabolism of glucose in 51 healthy adults. Neurology. 1995; 45:2077–2084.
52. Sultzer DL, Mahler ME, Cummings JL, Van Gorp WG, Hinkin CH, Brown C. Cor-
tical abnormalities associated with subcortical leisons in vascular dementia: clinical
and positron emission tomographic findings. Arch Neurol. 1995; 52:773–780.
53. Takahashi W, Takagi S, Ide M, Shohtsu A, Shinohara Y. Reduced cerebral glucose
metabolism in subjects with incidental hyperintensities on magnetic resonance
imaging. J Neurol Sci. 2000; 176:21–27.
54. Kennedy A, Frackowiak R, Newman S, Bloomfield PM, Seaward J, Rogues P,
Lewington G, Cunningham VJ, Rosser MN. Deficits in cerebral glucose metabo-
lism demonstrated by positron emission tomography in individuals at risk of
familial Alzheimer’s disease. Neurosci Lett. 1995; 186:1720.
55. Reiman E, Caselli R, Yun L, Chen K, Bandy D, Minoshima S, Thibodeau SN,
Osborne D. Preclinical evidence of Alzheimer’s disease in persons homozygous
for the E4 allele for apolipoprotein E. N Engl J Med. 1996; 334:752–758.
56. Reiman EM, Caselli RJ, Chen K, Alexander GE, Bandy D, Frost J. Declining brain
activity in cognitively normal apolipoprotein E varepsilon 4 heterozygotes: A
foundation for using positron emission tomography to efficiently test treatments
to prevent Alzheimer’s disease. Proc Nat Ac Sci USA. 2001; 98:3334–3339.
57. Ritchie K, Touchon J. Mild cognitive impairment: conceptual basis and current
nosological status. [see comments]. Lancet. 2000; 355 (9199):225–228.
58. Celsis P, Agniel A, Cardebat D, Demonet JF, Ousset PJ, Puel M. Age related cog-
nitive decline: a clinical entity? A longitudinal study of cerebral blood flow and
memory performance. J Neurol Neurosur Ps. 1997; 62:601–608.
136 THE AGEING BRAIN

59. McKelvey R, Bergman H, Stern J, Rush C, Zahirney G, Chertkow H. Lack of

prognostic significance of SPECT abnormalities in non-demented elderly subjects
with memory loss. Can J Neurol Sci. 1999; 26:23–28.
60. Small GW, La Rue A, Komo S, Kaplan A, Mandelkern MA. Predictors of cogni-
tive change in middle-aged and older adults with memory loss. Amer J Psych.
1995; 152:1757–1764.
61. Calautti C, Serrati C, Baron JC. Effects of age on brain activation during audi-
tory-cued thumb-to-index opposition — A positron emission tomography study.
Stroke. 2001; 32:139–146.
62. Ross MH, Yurgelun-Todd DA, Renshaw PF, Maas LC, Mendelson JH, Mello
NK, Gohen BM, Levin JM. Age-related reduction in functional MRI response to
photic stimulation. Neurology. 1997; 48:173–176.
63. Grady CL, McIntosh AR, Bookstein F, Horwitz B, Rapoport SI, Haxby JV. Age-
related changes in regional cerebral blood flow during working memory for faces.
Neuroimage. 1998; 8:409–425.
64. Madden DJ, Turkington TG, Provenzale JM, Hawk TC, Hoffman JM. Selective
and divided visual attention - age-related changes in regional cerebral blood flow
measured by (H2O)-O-15 pet. Hum Brain Mapp. 1997; 5:389–409.
65. Esposito G, Kirkby BS, Van Horn JD, Ellmore TM, Berman KF. Context-depend-
ent, neural system-specific neurophysiological concomitants of ageing: mapping
PET correlates during cognitive activation. Brain. 1999; 122:963–979.
66. Madden DJ, Turkington TG, Coleman RE, Provenzale JM, DeGrado TR, Hoff-
man JM. Adult age differences in regional cerebral blood flow during visual world
identification: evidence from H215O PET. Neuroimage. 1996; 3:127–142.
67. Gur RC, Gur RE, Obrist WD, Skolnick BE, Reivich M. Age and regional cerebral
blood flow at rest and during cognitive activity. Arch Gen Psychiat. 1987; 44:
617–621.
68. Grady CL, McIntosh AR, Horwitz B, Maisog JM, Ungerleider LG, Mentis MJ,
Pietrini P, Schapiro MB, Haxby JV. Age-related reductions in human recognition
memory due to impaired encoding. Science. 1995; 269 (5221):218–221.
69. Cabeza R, Grady CL, Nyberg L, McIntosh AR, Tulving E, Kapur S, Jennings
JM, Houle S, Craik FI. Age-related differences in neural activity during memory
encoding and retrieval: a positron emission tomography study. J Neurosci. 1997;
17:391–400.
70. Duara R, Gross-Glenn K, Barker WW, Chang JY, Apicella A, Loewenstein D,
Boothe T. Behavioral activation and the variability of cerebral glucose metabolic
measurements. J Cerebr Blood F Met. 1987; 7:266–271.
71. Skolnick BE, Gur RC, Stern MB, Hurtig HI. Reliability of regional cerebral blood
flow activation to cognitive tasks in elderly normal subjects. J Cerebr Blood F
Met. 1993; 13:448–453.
72. D’Esposito M, Zarahn E, Aguirre GK, Rypma B. The effect of normal aging on
the coupling of neural activity to the bold hemodynamic response. Neuroimage.
1999; 10:6–14.
73. Huettel SA, Singerman JD, McCarthy G. The effects of aging upon the hemo-
dynamic response measured by functional MRI. Neuroimage. 2001; 13:161–
175.
74. Levine BK, Beason-Held LL, Purpura KP, Aronchick DM, Optican LM, Alexan-
der GE, Horwitz B, Rapoport SI, Schapiro MB. Age-related differences in visual
perception: a PET study. Neurobiol Aging. 2000; 21:577–584.
75. Grady CL, Haxby JV, Horwitz B, Schapiro MB, Rapoport SI, Ungerleider LG,
Mishkin M, Carson RE, Herscovitch P. Dissociation of object and spatial vision
in human extrastriate cortex: Age-related changes in activation of regional cer-
ebral blood flow measured with [15O] water and positron emission tomography.
J Cognitive Neurosci. 1992; Vol 4:23–34.
CORRELATES OF BRAIN AGEING 137

76. Grady CL, Maisog JM, Horwitz B, Ungerleider LG, Mentis MJ, Salerno JA,
Pietrini P, Wagner E, Haxby JV. Age-related changes in cortical blood flow
activation during visual processing of faces and location. J Neurosci. 1994; 14:
1450–1462.
77. Grady CL, McIntosh AR, Horwitz B, Rapoport SI. Age-related changes in the
neural correlates of degraded and non-degraded face processing. Cognitive Neu-
ropsych. 2000; 17:165–186.
78. McIntosh AR, Sekuler AB, Penpeci C, Rajah MN, Grady CL, Sekuler R, Bennett
PJ. Recruitment of unique neural systems to suport visual memory in normal
aging. Curr Biol. 1999; 9:1275–1278.
79. Haut MW, Kuwabara H, Leach S, Callahan T. Age-related changes in neural
activation during working memory performance. Aging Neuropsychol C. 2000;
7:119–129.
80. Nagahama Y, Fukuyama H, Yamauchi H, Katsumi Y, Magata Y, Shibasaki H,
Kimura J. Age-related changes in cerebral blood flow activation during a Card
Sorting Test. Exp Brain Res. 1997; 114:571–577.
81. Herholz K, Ehlen P, Kessler J, Strotmann T, Kalbe E, Markowitsch HJ. Learning
face-name associations and the effect of age and performance: a PET activation
study. Neuropsychologia. 2001; 39:643–650.
82. Madden DJ, Turkington TG, Provenzale JM, Denny LL, Hawk TC, Gottlob LR,
Coleman RE. Adult age differences in the functional neuroanatomy of verbal
recognition memory. Hum Brain Mapp. 1999; 7:115–135.
83. Backman L, Almkvist O, Andersson J, Nordberg A, Winblad B, Reineck R, Lang-
strom B. Brain activation in young and older adults during implicit and explicit
retrieval. J Cognitive Neurosci. 1997; 9:391.
84. Hazlett EA, Buchsbaum MS, Mohs RC, Spiegel-Cohen J, Wei TC, Azueta R,
Haznedar MM, Singer MB, Shihabuddin L, Luu-Hsia C, Harvey PD. Age-related
shift in brain region activity during successful memory performance. Neurobiol
Aging. 1998; 19:437–445.
85. Cabeza R, Anderson ND, Houle S, Mangels JA, Nyberg L. Age-related differences
in neural activity during item and temporal-order memory retrieval: a positron
emission tomography study. J Cognitive Neurosci. 2000; 12:197–206.
86. Grachev ID, Apkarian AV. Aging alters regional multichemical profile of the
human brain: an in vivo 1H-MRS study of young versus middle-aged subjects.
J Neurochem. 2001; 76:582–593.
87. Kadota T, Horinouchi T, Kuroda C. Development and aging of the cerebrum: assess-
ment with proton MR spectroscopy. Amer J Neuroradiol. 2001; 22:128–135.
88. Charles HC, Lazeyras F, Krishman KR, Boyko OB, Patterson LJ, Doraiswamy
PM, McDonald WM. Proton Spectroscopy of human brain: effects of age and sex.
Prog Neuro-Psychoph. 1994; 18:995–1004.
89. Bruhn H, Stoppe G, Merboldt KD, Michaelis T, Hanicke W, Frahm J. Cerebral
metabolite alterations in normal aging and Alzheimer’s dementia (Abstract). Proc
Soc Magn Reson Med. 1992; 1:752.
90. Christiansen P, Toft P, Larsson HBW, Stubgaard M, Henriksen O. The Concen-
tration of N-acetyl aspartate, creatine + phosphocreatine, and choline in differ-
ent parts of the brain in adulthood and senium. Magn Reson Imaging. 1993; 11:
799–806.
91. Lim KO, Spielman DM. NAA in cortical gray matter with applications for meas-
uring changes due to aging. Magn Reson Med. 1997; 37:372–377.
92. Kreis R, Ernst T, Ross BD. Absolute quantitation of water and metabolites in the
human brain. II. Metabolite concentratons. J Magn Reson Imaging. 1993; 102:
9–19.
93. Soher BJ, van Zijl PCM, Duyn JH, Barker PB. Quantitative proton MR spectro-
scopic imaging of the human brain. Magn Reson Med. 1996; 35:356–363.
138 THE AGEING BRAIN

94. Chang L, Ernst T, Poland RE, Jenden DJ. In vivo proton magnetic resonance
spectroscopy of the normal aging human brain. Life Sci. 1996; 58:2049–2056.
95. Pfefferbaum A, Adalsteinsson E, Spielman D, Sullivan EV, Lim KO. In Vivo Spec-
troscopic Quantification of the N-acetyl moiety, creatine, and choline from large
volumes of brain gray and white matter: Effects of normal aging. Magn Reson
Med. 1999; 41:276–284.
96. Saunders DE, Howe FA, van den Boogaart A, Griffiths JR, Brown MM. Aging
of the adult human brain: in vivo quantitation of metabolite content with proton
magnetic resonance spectroscopy. J Magn Reson Imaging. 1999; 9:711–716.
97. Angelie E, Bonmartin A, Boudraa A, Gonnaud P-M, Mallet J-J, Sappey-Marinier
D. Regional differences and metabolic changes in normal aging of the human
brain: proton MR spectroscopic imaging study. Amer J Neuroradiol. 2001; 22:
119–127.
98. Christiansen PB, Toft P, Gideon ER, Danielsen PR, Henriksen O. MR-visible
water content in human brain: A proton MRS study. Magn Reson Imaging. 1994;
12:1237–1244.
Chapter 8

NEUROENDOCRINE
ASPECTS OF BRAIN
AGEING
George A Smythe

Introduction

Regulatory function of the human (and animal) body depends on the inte-
grated and co-ordinated activity of two major control systems: the endocrine
system and the nervous system. That there is a close interrelationship between
the function of the mind and endocrine hormone secretion is an old concept of
western medicine that came from findings such as the association of depres-
sion with dysfunction of the hypothalamic–pituitary–adrenal (HPA) axis.1
The field of neuroendocrinology arose out of observations pointing to signifi-
cant influences being exerted by hormones, and related peptides, on the brain
and vice versa. Research into this “neuroendocrine” hypothesis accelerated
from the early 1970s as the technologies became increasingly available for the
isolation, characterization and measurement of neurotransmitters, hypotha-
lamic peptides, pituitary and other endocrine hormones. Emerging techniques
are opening up new ways of examining brain chemistry; proton magnetic
spectroscopy, for example, has recently been used to show the marked reor-
ganization of brain chemical networks that occurs with normal ageing.2
Neuroendocrine interactions are critically important in normal human
development. The role of the brain, especially its neurotransmitters and
hypothalamic peptides, in control of the pituitary, thyroid, thymus, adrenals,
pancreas and gonads has been extensively documented from early to adult
development phases. Changes to these neuroendocrine systems post-maturity
and in the elderly are less well defined. The question arises as to whether there
are neuroendocrine factors which have roles in maintaining “healthy ageing,”
140 THE AGEING BRAIN

dysfunction of which may result in premature ageing and neurone loss. With
ageing there are notable declines in both mental and physical function that
may be mediated, in part at least, by known age-related changes in endocrine
function and feedback to the brain. These include:
• Cognitive function
• Reproductive function
• Muscle mass and strength
• Cardiac performance
• Immune function

Ageing and Endocrine Relationships

The major age-related changes in endocrine hormones include declines in

circulating levels and responses of pituitary growth hormone3–6 and gonadal
steroid hormones7–16, adrenal dehydroepiandrostenedione (DHEA) 17-24,
and insulin-like growth factor-1 (IGF-1). 13,25–28 On the other hand, it is
significant that secretion of pituitary ACTH and adrenal glucocorticoids (in
contrast to DHEA) trend upward with ageing.24,29-39 These latter findings
are consistent with significant changes to HPA function with ageing; evidence

Figure 1. Neuoendocrine aspects of ageing: Brain–body neuroendocrine feedback and

putative age-related changes to selected brain and endocrine hormones.
NEUROENDOCRINE ASPECTS OF BRAIN AGEING 141

of cortisol excess raises questions about the role, in ageing, of stress and its
effect on cognition and hippocampal neurons40-46.
In Figure 1 feedback of neural signals to the periphery and from the
peripheral target hormones (and products such as growth hormone-derived
IGF) to the pituitary and brain are indicated by the large shaded arrows.
The putative direction of change of neuroendocrine effectors with ageing are
indicated by the small arrows inside the boxes. Note 1, hypothalamic corti-
cotropin releasing hormone (CRH)42,47 and somatotropin release-inhibiting
factor (somatostatin, SRIF) increase;48 growth hormone releasing hormone
(GHRH) is decreased49,50 but downtrends in gonadotropin releasing hor-
mone (GnRH) show sexual dimorphism with changes being more evident in
females (data from animal studies — see below). Note 2, consistent with the
hypothalamic changes, there is evidence that pituitary secretion of ACTH is
increased and growth hormone (GH) is reduced but the data is less clear-cut
in the case of the pituitary gonadotropins where, again, there is evidence of
sexual dimorphism and significant variation between changes in luteinizing
hormone (LH) versus those in follicle stimulating hormone (FSH).51 Note 3,
as a consequence of reduced GH secretion, the product of its action on the
liver and other tissues,52 insulin-like growth factor-1 (IGF-1) is reduced. Note
4, the bulk of evidence is consistent with increased adrenal glucocorticoid
production whereas the adrenal androgenic steroid DHEA and its sulphate
is reduced.44 Here there is sexual dimorphism with women showing greater
changes than men with ageing.53 Note 5, estrogen levels decline following
menopause in women and testosterone levels are reduced in men (and women)
with ageing.51

Ageing, Stress, and the Hypothalamic–Pituitary–Adrenal (HPA) Axis

Evidence of altered HPA function with ageing comes from both animal and
human studies. Sapolsky and co-workers showed in the rat there is a significant
age-related increase in corticosterone production and that the ability of ani-
mals to “turn off” stress-induced glucocorticoid release is impaired.54,55 The
apparent age-related failure of glucocorticoid excess to exert normal feedback
inhibition on central, hypothalamic, or pituitary receptors has attracted con-
siderable research.24,33,34,40,42, 46,56–60 Normally, a principle brain response to
stress is markedly increased activity of noradrenergic neuronal activity. 61 This
increased noradrenergic drive acts on CRH neurons at the level of the paraven-
tricular nucleus of the hypothalamus and CRH is released into the pituitary
portal circulation to stimulate pituitary ACTH release (see Figure 2). Figure 2
also indicates the feedback of glucocorticoids at the level of the anterior pitui-
tary and certain brain centres to inhibit further CRF and ACTH release.
Figure 2 summarizes neuroendocrine control of the HPA axis and glu-
cocorticoid feedback. Included in this diagram are inputs to and from the
hippocampus, amygdala and the bed nucleus of the stria terminalis (BNST)
142 THE AGEING BRAIN

Figure 2. The hypothalamic-pituitary-adrenal-axis. Neuroendocrine control of

ACTH release and central sites of glucocorticoid feedback.

all of which can mediate inhibition or stimulation of CRH release.42,62,63

Using modern analytical methods, the hypothalamic neuronal activity of
norepinephrine (NE) and serotonin (5-HT) at terminals arising from cells
in the brainstem (locus coeruleus) and midbrain (Raphe) can be assessed by
measuring the transmitters and their primary neuronal metabolites (DHPG
and 5-HIAA, respectively).61,64

The “glucocorticoid cascade” hypothesis of ageing and hippocampal

damage
A number of animal studies have shown an age-related chronic increase in
corticosterone and an apparent failure of glucocorticoid negative feedback.
These data, taken with evidence that excess glucocorticoid levels caused dam-
age to hippocampal neurones that are involved in cognition, led Sapolsky et
al. to propose the glucocorticoid cascade hypothesis of ageing and hippoc-
ampal damage.65 In ageing man there is a decline in cognitive function and
several groups have proposed that this may be due to cortisol excess in accord
with Sapolsky’s hypothesis.40,41,57,66,67 Not all studies support the hypothesis
in man68,69 and Angelucci, in questioning earlier conclusions, has suggested
that changes seen represent an adaptive response to a CNS neurodegenerative
inflammatory process.69 It should be noted that, in man, age-related eleva-
NEUROENDOCRINE ASPECTS OF BRAIN AGEING 143

Figure 3. Effects of age on the HPA axis in male Wistar rats. Means± SEM are shown,
n=6 per group.

tions of cortisol are not always significant and there are sex differences.44,68
However most evidence does point to a failure of glucocorticoid feedback
with ageing but the level at which this failure occurs is not clear.29,31,35,36,70
The author has investigated the status of hypothalamic noradrenergic neuro-
nal activity, circulating ACTH and corticosterone in young (2 months old)
compared with old (20–24 months old) rats. The results of this unpublished
study are shown in Figure 3. Consistent with published results47 the data
show a significant increase in serum concentrations of ACTH and a non-
significant increase in serum corticosterone. The novel finding here is that
of a significant decrease in hypothalamic neuronal activity (DHPG/NE)61 in
the old rats. This is discordant with the HPA relationships in normal animals
subjected to stress where there is a positive relationship between ACTH and
medial basal hypothalamus (mbh) DHPG/NE ratio.61 These data indicate
that in the rat, with respect to the HPA axis, ageing is not akin to stress and
that the failure of feedback inhibition does not seem to occur at the level of
the hypothalamus.

Neuroendocrine Changes with the Menopause

Ovulatory failure is one of the earliest events of ageing and is one that clearly
involves neuroendocrine mechanisms. The central nervous system has been
described as the pacemaker of reproductive senescence.71 The consequences
of menopause for women relate not only to the loss of reproductive ability but
the loss of ovarian follicular estrogen and the decline of circulating estrogen
levels. In a major review, Wise et al.72 have highlighted many findings with
respect to neurotrophic and neuroprotective properties of estrogen. These
findings reinforce the ideas underlying estrogen replacement therapy,73,74 for
which there are negative as well as positive issues to be considered.75
The reported benefits of peri- and post-menopausal estrogen replacement
include maintenance of normal bone and mineral metabolism,76,77 memory
and cognition,78–82 decreased sympathetic nervous system activity, 83,84
144 THE AGEING BRAIN

decreased cortisol and HPA axis response.85-87 Improved IGF-1 production

is seen with transdermal estrogen administration compared with the oral
route;25 the transdermal route of estrogen administration is also associated
with improved insulin sensitivity and glucose metabolism.88
Improved memory and attention in postmenopausal women with Alzhe-
imer’s disease has been reported following estrogen administration.79 How-
ever, not all investigations have demonstrated a protective effect of estrogen
treatment against declines in cognitive function or stress reponses in older
non-demented women.89,90 Two recent randomized, double-blind trials of
estrogen treatment in women with Alzheimer’s disease (AD) failed to demon-
strate any cognitive or functional improvement following estogen.91,92 While
it is clear from animal and clinical studies that estrogen acts in the brain via
a number of established estrogen receptors as a neuromodulatory and neuro-
protective hormone,93-95 it does not appear to act to restore existing neural
damage.

The Age-Related Decline in Growth Hormone and IGF-1

Human growth hormone secretion (hGH) and IGF-1 production undergo

significant declines with ageing.96 The age-related decline in hGH with ageing
is exemplified by the comprehensive studies of van Cauter and colleagues into
sleep-related hGH release.39 Which neuroendocrine factors either mediate or
are affected by the age-related changes have not been established, but several
possibilities do arise (cf. Figure 1). These include: i) a primary pituitary defect,
ii) increased hypothalamic release of SRIF to inhibit pituitary GH release,
iii) reduced hypothalamic release of GHRH, and iv) altered activity of the
hypothalamic neurotransmitters that mediate release of SRIF and/or GHRH
— either primarily or as a consequence of reduced negative feedback.
The bulk of research data from both man and animals mitigates against a
primary pituitary defect and favours reduced GHRH activity being a major
contributor to the age-related decline in GH.6,49,97–102 Some evidence also
supports the possibility that there may also be increased SRIF activity with
ageing.49,50,103 Alterations in the activities of these peptides may, in turn, be
mediated by their controlling neurotransmitters. The monoaminergic control
of GHRH and SRIF has been a contentious issue for many years with both
catecholamines and serotonin being proposed as the primary mediators of
hypothalamic GHRH release.104–107 Direct measurement of hypothalamic
monoamine neurotransmitter activity in the rat is consistent with serotonin
neuronal activity being a major activator of GHRH release.105,106,108 When
considered with the reduced activity of GHRH noted above, it is conceivable
that there may be an age-related decline in central serotonin function and
support for this is found in human studies demonstrating reduced brain sero-
tonin activity and binding sites.109–112 If serotoninergic systems in the brain
are reduced with ageing then this would not appear to be due to reduced GH
NEUROENDOCRINE ASPECTS OF BRAIN AGEING 145

negative feedback. In the rat at least, GH lack (hypophysectomy) is associated

with increased hypothalamic serotonin neuronal activity which is reduced to
normal by exogenous GH administration.105
GH and IGF-1 decline with ageing and, taken with the ready availability
of recombinant GH, there has been considerable interest in the role of GH
replacement therapy to treat this so-called “somatopause”.113–119 This con-
cept is not without controversy and Morley120 has questioned whether, in this
context, “GH is a fountain of youth or a death hormone?” The general con-
sensus points to the need for carefully controlled studies.113,115,119,121–123

Androgens and Aging: the “Adrenopause”

LH secretion in men is regulated by release of GnRH from the pituitary and

negative feedback of circulating testosterone. With ageing, this neuroendo-
crine control is altered as LH levels increase at the rate of 1.9% per year,124
serum testosterone levels decline at the rate of about 0.4% per year123 and
DHEA decreases at a faster rate of 3.1%.124 On the basis of a relative “hypog-
onadism of ageing” in older men, hormone replacement with testosterone
and DHEA have been investigated. In general, testosterone replacement in
older hypogonadal males have indicated positive effects on muscle mass, bone
mineral density and fat mass but it is too early to establish its true efficacy51
or whether it has any effect in improving neural function.
In developing humans, DHEA and its sulphate (DHEAS) are the most
abundant steroids but their levels decline after adrenarche and with ageing.
DHEAS treatment in animals has been reported to have memory-enhancing
properties and replacement in ageing has been proposed.125, 126 Longitudi-
nal studies in elderly men and women have failed to show any association
between DHEA levels and cognitive performance127 and caution has been
advised in relation to its use as a hormone replacement in ageing.123, 128

Summary

Neuroendocrine systems change with ageing. The HPA axis is altered, partic-
ularly with respect to glucocorticoid responses and feedback to the pituitary
and brain. These changes are proposed to alter hippocampal neurons and
cognition. Menopause is associated with marked changes as the decline in
estrogen levels reaching the brain take effect. The case for estrogen replace-
ment is strong. Estrogen replacement appears able to prevent (but not repair)
brain neuronal damage but specific brain actions at the hippocampus and
other regions require further study. The ageing male undergoes a so-called
“adrenopause” with falling androgens and brain feedback but effects of exog-
enous androgen treatment is less well studied than that of the estrogen defi-
cient aged female. Age-related declines of GH and IGF-1 are associated with
146 THE AGEING BRAIN

physical and metabolic deterioration and, at the level of the brain, declining
serotoninergic systems. The change in this neurotransmitter may reflect a
primary failure in GHRH stimulation with ageing.
While many of the neuroendocrine changes seen with ageing are reminis-
cent of those seen with chronic stress, there is no clear evidence that ageing
equates with stress. No doubt, however, stress can accelerate or exacerbate
the effects of ageing.

Acknowledgements

The author wishes to express his sincere thanks to Mr Ray Williams for his
inspirational support and for generously providing equipment for our work.
I also wish to thank all of my colleagues in the BMSF for their constructive
assistance.

References

1. Carroll B, Mendels J. Neuroendocrine regulation in affective disorder. In: Sachar

E, editor. Hormones, behavior, and psychopathology. New York: Raven Press,
1976:193–224.
2. Grachev ID, Apkarian AV. Chemical network of the living human brain. Evidence
of reorganization with aging. Brain Res Cogn Brain Res. 2001; 11:185–197.
3. Gil-Ad I, Gurewitz R, Marcovici O, Rosenfeld J, Laron Z. Effect of aging on
human plasma growth hormone response to clonidine. Mech Ageing Dev. 1984;
27:97–100.
4. Muggeo M, Fedele D, Tiengo A, Molinari M, Crepaldi G. Human growth hor-
mone and cortisol response to insulin stimulation in aging. J Gerontol. 1975; 30:
546–551.
5. Rudman D, Kutner MH, Rogers CM, Lubin MF, Fleming GA, Bain RP. Impaired
growth hormone secretion in the adult population: relation to age and adiposity.
J Clin Invest. 1981; 67:1361–1369.
6. Russell-Aulet M, Dimaraki EV, Jaffe CA, DeMott-Friberg R, Barkan AL. Aging-
related growth hormone (GH) decrease is a selective hypothalamic GH-releasing
hormone pulse amplitude mediated phenomenon. J Gerontol A Biol Sci Med Sci.
2001; 56:M124–129.
7. Wise PM, Smith MJ, Dubal DB, Wilson ME, Krajnak KM, Rosewell KL. Neu-
roendocrine influences and repercussions of the menopause. Endocr Rev. 1999;
20:243–248.
8. Jiroutek MR, Chen MH, Johnston CC, Longcope C. Changes in reproductive hor-
mones and sex hormone-binding globulin in a group of postmenopausal women
measured over 10 years. Menopause. 1998; 5:90–94.
9. Santoro N, Banwell T, Tortoriello D, Lieman H, Adel T, Skurnick J. Effects of
aging and gonadal failure on the hypothalamic-pituitary axis in women. Am J
Obstet Gynecol. 1998; 178:732–741.
10. Coulam CB. Age, Estrogens, and the psyche. Clin Obstet Gynecol. 1981; 24:
219–229.
11. Basaria S, Dobs AS. Hypogonadism and androgen replacement therapy in elderly
men. Am J Med. 2001; 110:563–572.
NEUROENDOCRINE ASPECTS OF BRAIN AGEING 147

12. Harman SM, Metter EJ, Tobin JD, Pearson J, Blackman MR. Longitudinal effects
of aging on serum total and free testosterone levels in healthy men. Baltimore
Longitudinal Study of Aging. J Clin Endocr Metab. 2001; 86:724–731.
13. Leifke E, Gorenoi V, Wichers C, Von Zur Muhlen A, Von Buren E, Brabant G.
Age-related changes of serum sex hormones, insulin-like growth factor-1 and
sex-hormone binding globulin levels in men: cross-sectional data from a healthy
male cohort. Clin Endocrinol (Oxf). 2000; 53:689–695.
14. Nankin HR, Calkins JH. Decreased bioavailable testosterone in aging normal and
impotent men. J Clin Endocr Metab. 1986; 63:1418–1420.
15. Vermeulen A. Gonadal senescence in the male. J Endocr Invest. 1985; 8:93–98.
16. Noth RH, Mazzaferri EL. Age and the endocrine system. Clin Geriatr Med. 1985;
1:223–250.
17. Ohashi M, Kato K, Nawata H, Ibayashi H. Adrenocortical responsiveness to
graded ACTH infusions in normal young and elderly human subjects. Gerontol-
ogy. 1986; 32:43–51.
18. Pavlov EP, Harman SM, Chrousos GP, Loriaux DL, Blackman MR. Responses
of plasma adrenocorticotropin, cortisol, and dehydroepiandrosterone to ovine
corticotropin-releasing hormone in healthy aging men. J Clin Endocr Metab.
1986; 62:767–772.
19. Baulieu EE. Studies on dehydroepiandrosterone (DHEA) and its sulphate during
aging. CR Acad Sci III. 1995; 318:7–11.
20. Hornsby PJ. Biosynthesis of DHEAS by the human adrenal cortex and its age-
related decline. Ann NY Acad Sci. 1995; 774:29–46.
21. Hinson JP, Raven PW. DHEA deficiency syndrome: a new term for old age? J
Endocrinol. 1999; 163:1–5.
22. Parker CR, Jr. Dehydroepiandrosterone and dehydroepiandrosterone sulfate
production in the human adrenal during development and aging. Steroids. 1999;
64:640–647.
23. Parker CR, Jr., Slayden SM, Azziz R, et al. Effects of aging on adrenal function
in the human: responsiveness and sensitivity of adrenal androgens and cortisol
to adrenocorticotropin in premenopausal and postmenopausal women. J Clin
Endocr Metab. 2000; 85:48–54.
24. Ferrari E, Cravello L, Muzzoni B, et al. Age-related changes of the hypothalamic-
pituitary-adrenal axis: pathophysiological correlates. Eur J Endocrinol. 2001;
144:319–329.
25. Ho KK, O’Sullivan AJ, Weissberger AJ, Kelly JJ. Sex steroid regulation of growth
hormone secretion and action. Horm Res. 1996; 45:67–73.
26. Toogood AA, Shalet SM. Ageing and growth hormone status. Baillieres Clin
Endocrinol Metab. 1998; 12:281–296.
27. Aleman A, de Vries WR, de Haan EH, Verhaar HJ, Samson MM, Koppeschaar
HP. Age-sensitive cognitive function, growth hormone and insulin-like growth
factor 1 plasma levels in healthy older men. Neuropsychobiology. 2000; 41:
73–78.
28. Ravaglia G, Forti P, Maioli F, et al. Body composition, sex steroids, IGF-1, and
bone mineral status in aging men. J Gerontol A Biol Sci Med Sci. 2000; 55:
M516–521.
29. Born J, Ditschuneit I, Schreiber M, Dodt C, Fehm HL. Effects of age and gender
on pituitary-adrenocortical responsiveness in humans. Eur J Endocrinol. 1995;
132:705–711.
30. Copinschi G, Van Cauter E. Effects of ageing on modulation of hormonal secre-
tions by sleep and circadian rhythmicity. Horm Res. 1995; 43:20–24.
31. Ferrari E, Magri F, Dori D, et al. Neuroendocrine correlates of the aging brain in
humans. Neuroendocrinology. 1995; 61:464–470.
148 THE AGEING BRAIN

32. Raskind MA, Peskind ER, Pascualy M, et al. The effects of normal aging on
cortisol and adrenocorticotropin responses to hypertonic saline infusion. Psy-
choneuroendocrinology. 1995; 20:637–644.
33. Deuschle M, Gotthardt U, Schweiger U, et al. With aging in humans the activity
of the hypothalamus-pituitary-adrenal system increases and its diurnal amplitude
flattens. Life Sci. 1997; 61:2239–2246.
34. Magri F, Locatelli M, Balza G, et al. Changes in endocrine circadian rhythms as
markers of physiological and pathological brain aging. Chronobiol Int. 1997; 14:
385–396.
35. Wilkinson CW, Peskind ER, Raskind MA. Decreased hypothalamic-pituitary-
adrenal axis sensitivity to cortisol feedback inhibition in human aging. Neuroen-
docrinology. 1997; 65:79–90.
36. Boscaro M, Paoletta A, Scarpa E, et al. Age-related changes in glucocorticoid fast
feedback inhibition of adrenocorticotropin in man. J Clin Endocr Metab. 1998;
83:1380–1383.
37. Luisi S, Tonetti A, Bernardi F, et al. Effect of acute corticotropin releasing factor
on pituitary-adrenocortical responsiveness in elderly women and men. J Endocri-
nol Invest. 1998; 21:449–453.
38. Giordano R, Arvat E, Maccagno B, et al. Corticotroph and adrenal responsive-
ness to hCRH, hexarelin and ACTH in young and elderly subjects. J Endocrinol
Invest. 1999; 22:82.
39. Van Cauter E, Leproult R, Plat L. Age-related changes in slow wave sleep and
REM sleep and relationship with growth hormone and cortisol levels in healthy
men. J Amer Med Assoc. 2000; 284:861–868.
40. Lupien SJ, Nair NP, Briere S, et al. Increased cortisol levels and impaired cogni-
tion in human aging: implication for depression and dementia in later life. Rev
Neurosci. 1999; 10:117–139.
41. McEwen BS. Stress and the aging hippocampus. Front Neuroendocrin. 1999; 20:
49–70.
42. Pedersen WA, Wan R, Mattson MP. Impact of aging on stress-responsive neu-
roendocrine systems. Mech Ageing Dev. 2001; 122:963–983.
43. McEwen B, Chao H, Spencer R, Brinton R, Macisaac L, Harrelson A. Corticos-
teroid receptors in brain: relationship of receptors to effects in stress and aging.
Ann NY Acad Sci. 1987; 512:394–401.
44. Yen SS, Laughlin GA. Aging and the adrenal cortex. Exp Gerontol. 1998; 33:
897–910.
45. Bremner JD, Narayan M. The effects of stress on memory and the hippocampus
throughout the life cycle: implications for childhood development and aging. Dev
Psychopathol. 1998; 10:871–885.
46. Gotthardt U, Schweiger U, Fahrenberg J, Lauer CJ, Holsboer F, Heuser I. Cor-
tisol, ACTH, and cardiovascular response to a cognitive challenge paradigm in
aging and depression. Am J Physiol. 1995; 268:R865–873.
47. Herman JP, Larson BR, Speert DB, Seasholtz AF. Hypothalamo-pituitary-adreno-
cortical dysregulation in aging F344/Brown-Norway F1 hybrid rats. Neurobiol
Aging. 2001; 22:323–332.
48. Ghigo E, Arvat E, Gianotti L, et al. Human aging and the GH-IGF-I axis. J Pediatr
Endocr Met. 1996; 9:271–278.
49. Soule SG, Macfarlane P, Levitt NS, Millar RP. Contribution of growth hormone-
releasing hormone and somatostatin to decreased growth hormone secretion in
elderly men. S Afr Med J. 2001; 91:254–260.
50. Arvat E, Ceda GP, Di Vito L, et al. Age-related variations in the neuroendocrine
control, more than impaired receptor sensitivity, cause the reduction in the GH-
releasing activity of GHRPs in human aging. Pituitary. 1998; 1:51–58.
51. Morley JE. Androgens and aging. Maturitas. 2001; 38:61–71.
NEUROENDOCRINE ASPECTS OF BRAIN AGEING 149

52. Le Roith D, Scavo L, Butler A. What is the role of circulating IGF-I? Trends
Endocrin Met. 2001; 12:48–52.
53. Laughlin GA, Barrett-Connor E. Sexual dimorphism in the influence of advanced
aging on adrenal hormone levels: the Rancho Bernardo Study. J Clin Endocr Met.
2000; 85:3561–3568.
54. Sapolsky RM, Krey LC, McEwen BS. The adrenocortical stress-response in the
aged male rat: impairment of recovery from stress. Exp Gerontol. 1983; 18:
55–64.
55. Sapolsky RM. Glucocorticoids, stress, and their adverse neurological effects:
relevance to aging. Exp Gerontol. 1999; 34:721–732.
56. Cizza G, Gold PW, Chrousos GP. Aging is associated in the 344/N Fischer rat
with decreased stress responsivity of central and peripheral catecholaminergic
systems and impairment of the hypothalamic-pituitary-adrenal axis. Ann NY
Acad Sci. 1995; 771:491–511.
57. Wang PS, Lo MJ, Kau MM. Glucocorticoids and aging. J Formos Med Assoc.
1997; 96:792–801.
58. De Kloet ER, Sutanto W, Rots N, et al. Plasticity and function of brain corticos-
teroid receptors during aging. Acta Endocrinol–(Cop.) 1991; 125:65–72.
59. Heuser IJ, Gotthardt U, Schweiger U, et al. Age-associated changes of pituitary-
adrenocortical hormone regulation in humans: importance of gender. Neurobiol
Aging. 1994; 15:227–231.
60. Dodt C, Theine KJ, Uthgenannt D, Born J, Fehm HL. Basal secretory activity of
the hypothalamo-pituitary-adrenocortical axis is enhanced in healthy elderly. An
assessment during undisturbed night-time sleep. Eur J Endocrinol. 1994; 131:
443–450.
61. Smythe GA, Bradshaw JE, Vining RF. Hypothalamic monoamine control of
stress-induced adrenocorticotropin release in the rat. Endocrinology. 1983; 113:
1062–1071.
62. Feldman S, Conforti N, Weidenfeld J. Limbic pathways and hypothalamic neu-
rotransmitters mediating adrenocortical responses to neural stimuli. Neurosci
Biobehav Rev. 1995; 19:235–240.
63. Lee Y, Davis M. Role of the hippocampus, the bed nucleus of the stria terminalis,
and the amygdala in the excitatory effect of corticotropin-releasing hormone on
the acoustic startle reflex. J Neurosci. 1997; 17:6434–6446.
64. Smythe GA, Brandstater JF, Lazarus L. Serotoninergic control of rat growth
hormone secretion. Neuroendocrinology. 1975; 17:245–257.
65. Sapolsky RM, Krey LC, McEwen BS. The neuroendocrinology of stress and aging:
the glucocorticoid cascade hypothesis. Endocrinol Rev. 1986; 7:284–301.
66. Martignoni E, Costa A, Sinforiani E, et al. The brain as a target for adrenocortical
steroids: cognitive implications. Psychoneuroendocrinology. 1992; 17:343–354.
67. O’Brien JT. The ‘glucocorticoid cascade’ hypothesis in man: prolonged stress may
cause permanent brain damage. Br J Psychiat. 1997; 170:199–201.
68. Kudielka BM, Schmidt-Reinwald AK, Hellhammer DH, Schurmeyer T, Kirsch-
baum C. Psychosocial stress and HPA functioning: no evidence for a reduced
resilience in healthy elderly men. Stress. 2000; 3:229–240.
69. Angelucci L. The glucocorticoid hormone: from pedestal to dust and back. Eur J
Pharmacol. 2000; 405:139–147.
70. Veldhuis JD, Iranmanesh A, Samojlik E, Urban RJ. Differential sex steroid nega-
tive feedback regulation of pulsatile follicle-stimulating hormone secretion in
healthy older men: deconvolution analysis and steady-state sex-steroid hormone
infusions in frequently sampled healthy older individuals. J Clin Endocr Met.
1997; 82:1248–1254.
71. Wise PM. Neuroendocrine modulation of the “menopause”: insights into the
aging brain. Am J Physiol. 1999; 277:E965–970.
150 THE AGEING BRAIN

72. Wise PM, Dubal DB, Wilson ME, Rau SW, Liu Y. Estrogens: trophic and protec-
tive factors in the adult brain. Front Neuroendocrin. 2001; 22:33–66.
73. Lichtman R. Perimenopausal hormone replacement therapy. Review of the litera-
ture. J Nurse Midwifery. 1991; 36:30–48.
74. Bjorntorp P. Neuroendocrine ageing. J Intern Med. 1995; 238:401–404.
75. Nerhood RC. Making a decision about ERT/HRT. Evidence to consider in ini-
tiating and continuing protective therapy. Postgrad Med J. 2001; 109:167–170,
173–174, 178.
76. Kulak CA, Bilezikian JP. Osteoporosis: preventive strategies. Int J Fertil Womens
M. 1998; 43:56–64.
77. Shiflett S, Cooke CE. Osteoporosis: a focus on treatment. Maryland State Med J.
1997; 46:303–307.
78. Verghese J, Kuslansky G, Katz MJ, et al. Cognitive performance in surgically
menopausal women on estrogen. Neurology. 2000; 55:872–874.
79. Erkkola R. Female menopause, hormone replacement therapy, and cognitive
processes. Maturitas. 1996; 23:S27–30.
80. Asthana S, Craft S, Baker LD, et al. Cognitive and neuroendocrine response to
transdermal estrogen in postmenopausal women with Alzheimer’s disease: results
of a placebo-controlled, double-blind, pilot study. Psychoneuroendocrinology.
1999; 24:657–677.
81. LeBlanc ES, Janowsky J, Chan BK, Nelson HD. Hormone replacement therapy
and cognition: systematic review and meta-analysis. J Amer Med Assoc. 2001;
285:1489–1499.
82. Rice MM, Graves AB, McCurry SM, et al. Postmenopausal estrogen and estrogen-
progestin use and 2-year rate of cognitive change in a cohort of older Japanese
American women: The Kame Project. Arch Intern Med. 2000; 160:1641–1649.
83. Menozzi R, Cagnacci A, Zanni AL, Bondi M, Volpe A, Del Rio G. Sympathoad-
renal response of postmenopausal women prior and during prolonged administra-
tion of estradiol. Maturitas. 2000; 34:275–281.
84. Ceresini G, Freddi M, Izzo S, et al. Post-menopausal estrogen supplementation
only partially blunts the sympathoadrenal response to mental stress. J Endocrinol
Invest. 1999; 22:72–73.
85. Prinz P, Bailey S, Moe K, Wilkinson C, Scanlan J. Urinary free cortisol and sleep
under baseline and stressed conditions in healthy senior women: effects of estro-
gen replacement therapy. J Sleep Res. 2001; 10:19–26.
86. Lindheim SR, Legro RS, Bernstein L, et al. Behavioral stress responses in pre-
menopausal and postmenopausal women and the effects of estrogen. Am J Obstet
Gynecol. 1992; 167:1831–1836.
87. Kudielka BM, Schmidt-Reinwald AK, Hellhammer DH, Kirschbaum C. Psy-
chological and endocrine responses to psychosocial stress and dexamethasone/
corticotropin-releasing hormone in healthy postmenopausal women and young
controls: the impact of age and a two-week estradiol treatment. Neuroendocrinol-
ogy. 1999; 70:422–430.
88. O’Sullivan AJ, Ho KK. A comparison of the effects of oral and transdermal estro-
gen replacement on insulin sensitivity in postmenopausal women. J Clin Endocr
Met. 1995; 80:1783–1788.
89. Matthews KA, Flory JD, Owens JF, Harris KF, Berga SL. Influence of estrogen
replacement therapy on cardiovascular responses to stress of healthy postmeno-
pausal women. Psychophysiology. 2001; 38:391–398.
90. Matthews K, Cauley J, Yaffe K, Zmuda JM. Estrogen replacement therapy and cogni-
tive decline in older community women. J Am Geriatr Soc. 1999; 47:518–523.
91. Mulnard RA, Cotman CW, Kawas C, et al. Estrogen replacement therapy for
treatment of mild to moderate Alzheimer disease: a randomized controlled trial.
Alzheimer’s Disease Cooperative Study. J Amer Med Assoc. 2000; 283:1007–
1015.
NEUROENDOCRINE ASPECTS OF BRAIN AGEING 151

92. Henderson VW, Paganini-Hill A, Miller BL, et al. Estrogen for Alzheimer’s dis-
ease in women: randomized, double-blind, placebo-controlled trial. Neurology.
2000; 54:295–301.
93. Garcia-Segura LM, Azcoitia I, DonCarlos LL. Neuroprotection by estradiol.
Prog Neurobiol. 2001; 63:29–60.
94. Shughrue PJ, Merchenthaler I. Evidence for novel estrogen binding sites in the
rat hippocampus. Neuroscience. 2000; 99:605–612.
95. Dubal DB, Zhu H, Yu J, et al. Estrogen receptor alpha, not beta, is a critical link
in estradiol-mediated protection against brain injury. Proc Natl Acad Sci USA.
2001; 98:1952–1957.
96. Corpas E, Harman SM, Blackman MR. Human growth hormone and human
aging. Endocr Rev. 1993; 14:20–39.
97. Corpas E, Harman SM, Pineyro MA, Roberson R, Blackman MR. Growth hor-
mone (GH)-releasing hormone-(1–29) twice daily reverses the decreased GH and
insulin-like growth factor-I levels in old men. J Clin Endocrinol Metab. 1992;
75:530–535.
98. Merriam GR, Buchner DM, Prinz PN, Schwartz RS, Vitiello MV. Potential
applications of GH secretagogs in the evaluation and treatment of the age-
related decline in growth hormone secretion. Endocrine. 1997; 7:49–52.
99. Guldner J, Schier T, Friess E, Colla M, Holsboer F, Steiger A. Reduced efficacy
of growth hormone-releasing hormone in modulating sleep endocrine activity
in the elderly. Neurobiol Aging. 1997; 18:491–495.
100. degli Uberti EC, Ambrosio MR, Cella SG, et al. Defective hypothalamic growth
hormone (GH)-releasing hormone activity may contribute to declining GH
secretion with age in man. J Clin Endocrin Metab. 1997; 82:2885–2888.
101. Mulligan T, Jaen-Vinuales A, Godschalk M, Iranmanesh A, Veldhuis JD. Syn-
thetic somatostatin analog (octreotide) suppresses daytime growth hormone
secretion equivalently in young and older men: preserved pituitary respon-
siveness to somatostatin’s inhibition in aging. J Am Geriatr Soc. 1999; 47:
1422–1424.
102. Thorner MO, Chapman IM, Gaylinn BD, Pezzoli SS, Hartman ML. Growth hor-
mone-releasing hormone and growth hormone-releasing peptide as therapeutic
agents to enhance growth hormone secretion in disease and aging. Recent Prog
Horm Res. 1997; 52:215–244; (discussion) 244–246.
103. Marcell TJ, Wiswell RA, Hawkins SA, Tarpenning KM. Age-related blunting
of growth hormone secretion during exercise may not be soley due to increased
somatostatin tone. Metabolism. 1999; 48:665–670.
104. Muller EE. Some aspects of the neurotransmitter control of anterior pituitary
function. Pharmacol Res. 1989; 21:75–85.
105. Smythe GA, Duncan MW, Bradshaw JE, Cai WY. Serotoninergic control of
growth hormone secretion: hypothalamic dopamine, norepinephrine, and serot-
onin levels and metabolism in three hyposomatotropic rat models and in normal
rats. Endocrinology. 1982; 110:376–383.
106. Smythe GA, Gleeson RM, Stead BH. Stimulation of the hypothalamic-pituitary-
adrenal axis and inhibition of growth hormone release via increased central
noradrenaline neuronal activity by urethane anaesthesia in the rat: blockade by
clonidine. Aust J Biol Sci. 1987; 40:91–96.
107. Conway S, Richardson L, Speciale S, Moherek R, Mauceri H, Krulich L. Inter-
action between norepinephrine and serotonin in the neuroendocrine control of
growth hormone release in the rat. Endocrinology. 1990; 126:1022–1030.
108. Gotoh M, Hirooka Y, Tajima T, Iguchi A, Smythe GA. Adrenocorticotropin
and growth hormone secretions after intracerebroventricular administration of
neostigmine in rats: their relationships to hypothalamic monoaminergic neuro-
nal activities. Brain Res. 1994; 659:259–262.
152 THE AGEING BRAIN

109. Lerer B, Gelfin Y, Shapira B. Neuroendocrine evidence for age-related decline in

central serotonergic function. Neuropsychopharmacology. 1999; 21:321–322.
110. Kakiuchi T, Nishiyama S, Sato K, Ohba H, Nakanishi S, Tsukada H. Age-related
reduction of [11C]MDL100,907 binding to central 5-HT(2A) receptors: PET
study in the conscious monkey brain. Brain Res. 2000; 883:135–142.
111. van Dyck CH, Malison RT, Seibyl JP, et al. Age-related decline in central sero-
tonin transporter availability with [(123)I]beta-CIT SPECT. Neurobiol Aging.
2000; 21:497–501.
112. Meltzer CC, Smith G, DeKosky ST, et al. Serotonin in aging, late-life depression,
and Alzheimer’s disease: the emerging role of functional imaging. Neuropsy-
chopharmacology. 1998; 18:407–430.
113. Lamberts SW. The somatopause: to treat or not to treat? Horm Res. 2000; 53:
42–43.
114. Savine R, Sonksen P. Growth hormone – hormone replacement for the somato-
pause? Horm Res. 2000; 53:37–41.
115. Cummings DE, Merriam GR. Age-related changes in growth hormone secretion:
should the somatopause be treated? Semin Reprod Endocr 1999; 17:311–325.
116. Hoffman AR, Ceda GP. Should we treat the somatopause? J Endocr Invest.
1999; 22:4–6.
117. Lieberman SA, Hoffman AR. The somatopause: should growth hormone defi-
ciency in older people Be treated? Clin Geriatr Med. 1997; 13:671–684.
118. Hoffman AR, Lieberman SA, Butterfield G, et al. Functional consequences of
the somatopause and its treatment. Endocrine. 1997; 7:73–76.
119. Sonsken P. Growth hormone and the somatopause. Growth Horm IGF Res.
1999; 9:1–2.
120. Morley JE. Growth hormone: fountain of youth or death hormone? J Am Geri-
atr Soc. 1999; 47:1475–1476.
121. von Werder K. The somatopause is no indication for growth hormone therapy.
J Endocrinol Invest. 1999; 22:137–141.
122. Toogood AA, Shalet SM. Conflicts with the somatopause. Growth Horm IGF
Res. 1998; 8:47–54.
123. Janssens H, Vanderschueren DM. Endocrinological aspects of aging in men: is
hormone replacement of benefit? Eur J Obstet Gyn R B. 2000; 92:7–12.
124. Gray A, Feldman HA, McKinlay JB, Longcope C. Age, disease, and changing
sex hormone levels in middle-aged men: results of the Massachusetts Male Aging
Study. J Clin Endocr Metab. 1991; 73:1016–1025.
125. Baulieu EE. Dehydroepiandrosterone (DHEA): a fountain of youth? J Clin
Endocrin Metab. 1996; 81:3147–3151.
126. Majewska MD. Neuronal actions of dehydroepiandrosterone. Possible roles in
brain development, aging, memory, and affect. Ann NY Acad Sci. 1995; 774:
111–120.
127. Carlson LE, Sherwin BB. Relationships among cortisol (CRT), dehydroepian-
drosterone-sulfate (DHEAS), and memory in a longitudinal study of healthy
elderly men and women. Neurobiol Aging. 1999; 20:315–324.
128. Steel N. Dehydro-epiandrosterone and ageing. Age Ageing. 1999; 28:89–91.
Chapter 9

CEREBROVASCULAR
SYSTEM AND THE
AGEING BRAIN
Velandai K. Srikanth and
Geoffrey A. Donnan*

Introduction

It is inevitable that the vascular system of the brain undergoes change with
increasing age. The primary purpose of this chapter is to review the existing
literature with respect to the impact of ageing on anatomical and physiologi-
cal aspects of the cerebrovascular system. The links and postulated mecha-
nisms between ageing, cerebrovascular changes and disease states will also be
discussed, as this is of primary interest to the clinician.

Ageing and Cerebrovascular Anatomy

Ageing and cerebral macrovasculature

Ageing and Arterial Changes

The arterial tree supplying the brain is comprised of the extracranial large
arteries and the intracranial anastomotic network (The Circle of Willis),
together with the progressively smaller vessels that penetrate brain tissue
and supply specific areas of the brain. The defining feature of these groups
of blood vessels is that they contain smooth muscle and are considered to be
responsible for maintaining adequate cerebral blood flow.
*To whom correspondence should be addressed.
154 THE AGEING BRAIN

Arterial changes in ageing closely resemble changes seen in the vascular

tree elsewhere. The changes most often observed with ageing include intimal
thickening, medial fibrosis and loss of elasticity for the larger arterioles and
arteries. These changes may occur at a slower rate in cerebral arteries than
in peripheral arteries such as the radial artery or the coronary artery.1 The
magnitude of these changes is observed to increase with each decade from the
age of 55.2 Smaller intraparenchymal vessels often tend to show tortuosity,
loops and kinks.3 It has been postulated that these loops and kinks may have
been mistaken for the so-called Charcot-Bouchard aneurysms, given that cur-
rent pathological techniques are superior in delineating their characteristics.

Ageing, disease and arterial changes

Atherosclerosis is best considered as an ageing-related disease state that is
almost ubiquitously present in older humans. The longer one lives, the chance
of developing atherosclerotic disease increases. However, it is unclear as to
how much the pathogenesis of atherosclerosis is dependent on a biological
ageing process. Age-related cellular changes in the arterial wall might con-
tribute in part to the development of the atherosclerotic plaque over time. A
number of putative factors may lead to this including decreased ability of the
endothelium to repair in the presence of injury, altered control of vascular
smooth muscle proliferation and the interaction between smooth muscle and
circulating lipoproteins. However, increasing age is only one of many impor-
tant risk factors for the development of atherosclerosis. Changes in other
cardiovascular risk factors invariably occur with ageing, and make a major
contribution to the accelerated formation of atherosclerotic plaque.
Occlusive disease due to atherosclerosis is more frequent at the origin of
the internal carotid artery, the carotid siphon, the proximal middle cerebral
artery, the proximal anterior carotid artery and the proximal basilar artery.
Artery-to-artery embolization occurs as an end result of severe atherosclero-
sis of these vessels leading to well-known stroke syndromes. The increasing
incidence of stroke with age is thus a function of an ageing vascular system
with both intrinsic and extrinsic biological mechanisms at play.
Another important effect of age-related arterial changes and disease such
as atherosclerosis is a loss of elasticity and distensibility of vessels. This may
lead to decreased arterial compliance predisposing the affected individual to
systolic hypertensive disease. The potentially serious effects of systolic hyper-
tension on the elderly brain include stroke, white matter disease and cognitive
decline.

Ageing and cerebral microvasculature

Although the larger vessels described previously are responsible for main-
taining overall cerebral blood flow, the cerebral microvasculature is chiefly
responsible for the important function of providing active nutrient substrate
for the neuron. There is a good correlation between the extent of this capil-
lary network and the activity of functioning neurons in the cerebral cortex.4
CEREBROVASCULAR SYSTEM AND THE AGEING BRAIN 155

These microvessels have structural properties that provide the basis for the
Blood–Brain Barrier (BBB). The BBB is comprised of a continuous endothe-
lium lacking fenestrations, thus acting as a very selective barrier to blood-
borne substances reaching the central nervous system. Pinocytotic vesicles
are found in very low numbers in the endothelium indicating that the BBB is
comparatively less permeable than other organ endothelium. The BBB allows
the transport of lipid-soluble and water-soluble substances either by passive
(diffusion) or active transport. The cellular components of the BBB possess
specialized carrier processes to provide adequate transport of nutrients, hor-
mones and neurotransmitter peptides.

Ageing and blood–brain barrier

A number of changes in the Blood–Brain Barrier with the ageing process in
animal and human models have been described. Thickening of the basement
membrane in the ageing rat 5 together with other changes including loss of
capillary endothelial cells and elongation of remaining endothelial cells 6 have
been observed. A reduction of mitochondrial numbers has also been described
in the endothelial cells of the BBB in the monkey but not the rat, leading the
authors to postulate that some BBB changes with ageing may be species spe-
cific.6,7 In the human brain, examination of biopsy material showed thinning
of white matter capillaries with age possibly related to loss of pericytes and
thinning of endothelial cytoplasm.8 There is no evidence to date in humans
that endothelial mitochondrial numbers decline with age or that significant
changes occur in membrane permeability characteristics such as junctional
gaps and pinocytotic vesicle density.
In the rat model, alterations in BBB transport function include a decrease
in BBB choline transport with ageing and decreased brain glucose influx.6
However there is no conclusive evidence in the literature that age-related
permeability of the BBB is altered significantly in the absence of neurological
or vascular disease.
Although most reports have concentrated on changes in the afferent micro-
vasculature (arterioles and capillaries) in ageing, few have described specific
age-related changes in the efferent microvasculature (venules). These include
the observation of a non-inflammatory mural thickening of the venular sys-
tem due to collagenosis, predominantly in the periventricular region.9

Ageing, disease and cerebral microvasculature

The cerebral microvasculature has been implicated in diseases such as Alzhe-
imer’s disease (AD), cerebral amyloid angiopathy and potential disease states
such as leukoaraiosis (deep white matter change). However, the role of the
microvessels and the BBB in these disorders is a matter of ongoing research
and controversy. It remains difficult to separate out the effects of intrinsic
ageing and disease on microvascular changes.
Alzheimer’s disease increases in prevalence with every decade of life in the
elderly population. Intense debate and research continues in the search for
156 THE AGEING BRAIN

the aetiological factors responsible for the development of the disease. It is

tempting to consider that age-related changes to cerebral microvessels may
have some role to play in the pathogenesis of AD. It has been proposed that
a breakdown of BBB may be an essential step in the pathogenesis of AD.10
Several investigators have hypothesised that microvascular injury secondary
to amyloid deposition leads to leakage of substances that may cause neuronal
injury.11–14 However, other investigators have demonstrated a lack of BBB
alteration in either AD or non-demented controls by immunohistochemical.15
Similarly, in a post-mortem immunohistochemical study of AD compared to
vascular dementia, age-matched controls, other neurodegenerative disorders
and young controls, albumin leakage in the neuropil was demonstrated in all
groups with no statistically significant difference between groups.16 These
authors were also unable to consistently detect significant amounts of other
serum proteins in the neuropil such as IgG and complement C3c. They con-
cluded that alteration in BBB was neither a primary nor a consistent event in
AD. In summary, in spite of ongoing interest in the field, it is still undecided
whether BBB alterations are causally related to AD or merely an epiphenom-
enon.
Amyloid Angiopathy: The microvasculature of the ageing brain is suscepti-
ble to the development of a specific disease termed cerebral amyloid angiopa-
thy (CAA). This is characterized by the deposition of fibrillar amyloid mate-
rial in the arteriolar media, gradually replacing the smooth muscle component
of the arteriole. This leads to a weakened arterial wall that is susceptible to
BBB dysfunction, as well as rupture leading to lobar haemorrhages. The depo-
sition of vascular amyloid is extensive in cases of Alzheimer’s dementia, but is
also associated with other conditions such as radiation injury and vasculitis.
Amyloid deposition in CAA may lead to small brain infarcts due to micro-
vascular occlusion, and thus may contribute to a dementing syndrome.17 The
study of CAA in familial forms of the disease may provide important insights
into the mechanistic links between ageing, microvascular pathology and dis-
ease states of the brain.
Leukoaraiosis is the radiological finding of white matter changes that
are commonly visible in CT scans or MRI of brains of elderly people. These
changes may contribute to a small extent to age-related decline in intellectual
function.18 Some researchers have proposed that these changes may be related
to disease processes such as chronic hypertension, diabetes mellitus or even
AD.19 They hypothesize that these disease processes may lead to disruption
of the BBB, leading to protein leakage into white matter. Others implicate
“periventricular venous collagenosis” in the pathogenesis of this condition
on finding that venous changes were correlated with the presence of white
matter hyperintensity on magnetic resonance imaging, whereas only mild age-
related change was observed in the afferent micro-vessels.20 Their findings
led them to hypothesise that deep venous occlusion as an age-related process
may lead to a long-term increase in vascular resistance and insidious oedema
in surrounding tissue, contributing in part to the development of white matter
CEREBROVASCULAR SYSTEM AND THE AGEING BRAIN 157

lesions. However, these findings have not yet been duplicated in other studies.
Investigators who have used gadolinium-MRI and diffusion MRI techniques
have provided conflicting results regarding a role for the BBB in pathogenesis
of white matter lesions, although the methods used are qualitatively differ-
ent.21,22

Ageing and Cerebrovascular Physiology

Cerebral Blood Flow and metabolism — physiological concepts

Cerebral Blood Flow (CBF)

Great advances have been made into the research of CBF due to refinement in
existing techniques of measurement including positron emission tomography
(PET), 133Xenon inhalation computed tomography (Xenon-CT) and single
positron emission computed tomography (SPECT), and the development
of new techniques such as perfusion/diffusion magnetic resonance imaging
(DWI/PWI MR), magnetic resonance spectroscopy (MRS) and transcranial
Doppler.

Determinants of Cerebral Blood Flow

Average CBF is around 60 ml/min/100 g of brain tissue in the resting adult
under physiological conditions with the cerebral grey matter receiving
the bulk of supply. Gender differences have been described in global CBF,

Intracranial Pressure

vein
artery Arteriolar bed

Auto-regulation Synaptic Activity

Figure 1. Schematic summary of factors involved in regulating Cerebral Blood Flow

(CBF).
158 THE AGEING BRAIN

with women of pre-menopausal age showing higher levels than similar aged
men.23,24 This difference begins to disappear after the fourth decade of life,
suggesting a possible role for oestrogen in augmenting CBF.
CBF is directly related to cerebral perfusion pressure and inversely related
to cerebral vascular resistance (Figure 1). Regional CBF is also determined by
the level of synaptic activity and consequently regional cerebral metabolism
(refer section on cerebral metabolism). Other factors playing an important
role in regulation of CBF include intracranial pressure (ICP) and blood viscos-
ity.25

Cerebral perfusion pressure

Cerebral perfusion pressure is the difference between arterial inflow pressure
and venous outflow pressure and represents the driving pressure for CBF.
The mean perfusion pressure in humans varies between 50 to 150 mm Hg
(autoregulatory range) without affecting cerebral blood flow due to the phe-
nomenon of cerebral autoregulation (Figure 2).

Cerebral vascular resistance

Cerebral vascular resistance (CVR) is primarily a function of the vessel radius
such that even small changes in luminal diameter (via constriction or dilata-
tion) can have major effects on resistance. The larger extracranial vessels and
the intracranial pial vessels contribute to about half the total cerebrovascular
resistance. CVR is controlled largely by cerebral autoregulatory mechanisms
such that an increase in CVR leads to a reduction in CBF. Vascular resist-
ance also depends to a certain extent on the viscosity of blood. The factors
CBF (ml/100 g/min)

Figure 2. Idealized cerebral pressure–flow curve.

CEREBROVASCULAR SYSTEM AND THE AGEING BRAIN 159

that may play an important role in determining blood viscosity include

shear (flow) rate, red cell rheology, platelet aggregability, plasma protein
(fibrinogen) levels. An increase in blood viscosity leads to reduction in CBF.26
Reduction in blood flow may lead to further increase in viscosity leading to
a vicious cycle of events.

Cerebral autoregulation
Cerebral autoregulation is achieved by the tight coupling of cerebral per-
fusion pressure and the diameter responses of the arteriolar system in an
attempt to stabilise CBF in the presence of fluctuations in systemic arterial
pressure. Within the autoregulatory range, CBF remains constant due to
cerebral vasoconstriction when perfusion pressure increases as a result of
systemic arterial hypertension.
Neural, metabolic, myogenic and endothelial mechanisms have been
postulated to explain the phenomenon of autoregulation.27 It appears that
the autoregulation of pial vessels may be predominantly due to neurogenic
control whereas the intracerebral vessels may well be under the influence
of local metabolic phenomena such as changes in arterial CO 2 tension.
Autoregulatory control of cerebral pial vessels may be under the influence
of adrenergic and trigeminovascular systems although this is the subject of
further study.28,29 Parasympathetic innervation of cerebral blood vessels has
assumed greater importance in the last decade, contributing to neural vasodi-
lation by means of nitric oxide, acetylcholine, glutamate and other agonists.
However, investigators using animal models have not provided evidence for
a major role for this system in resting and autoregulatory control of CBF.
Smaller intracerebral vessels are extremely sensitive to metabolic influences
such as arterial pCO2 tension. CBF may change by up to 5% with every
mmHg change in arterial pCO2 within physiological ranges (30–60 mmHg).
Arterial response to CO2 may be lessened in situations such as hypotension
or cerebral ischaemia.30 The effect of CO2 is mediated by the autonomic sup-
ply of the vessels and perivascular cerebrospinal fluid pH. Arterial pO2 and
Hydrogen ion also play a role in CBF regulation. The myogenic theory holds
that changes in transmural pressure may directly affect the tone of the vas-
cular smooth muscle leading to vasoconstriction or vasodilation in response
to increase or decrease in systemic arterial pressure respectively.27,31 More
recently, investigators have shown that such smooth muscle response to pres-
sure changes may be dependent on the presence of intact endothelium.32 This
may be linked to the release of contractile substances from the endothelium
rather than a tonic inhibition of endothelium-derived relaxing factor (EDRF
or Nitric Oxide).33

Intracranial pressure
Intracranial pressure (ICP) is an important determinant of cerebral perfusion.
A rise in ICP leads to compression of intracerebral vessels causing a decrease
in perfusion. Conversely, a fall in ICP leads to an increase in cerebral per-
160 THE AGEING BRAIN

fusion. This mechanism appears to be important in maintaining relatively

constant cerebral perfusion in states of positive and negative “g” and physi-
ological states of straining. Extreme rises in ICP (> 30 mm Hg) can lead to
stimulation of the brainstem vasomotor centre, leading to reflex increase in
systemic arterial pressure in order to maintain adequate cerebral perfusion.

Cerebral metabolism
The brain has a high demand for oxygen and glucose in order to maintain
optimal neuronal function even at rest. There are a number of unique features
regarding the activity and metabolism of the brain. More than a century
ago, it was postulated that cerebral metabolism was closely coupled with the
level of pre-synaptic activity in the brain.34 Results of studies using various
techniques appear largely to support this hypothesis.35–38 This coupling may
(hypothetically) occur as a result of a glutamate-mediated uptake of glucose
into peri-synaptic astrocytes leading to lactate production. Lactate can then
be used as energy substrate in the pre-synaptic vesicle. Cerebral metabolism
and CBF are also closely coupled with functional neuronal activity as an
intermediate link, suggesting that cerebral metabolism may be an important
determinant of CBF.34,39,40 This coupling may be mediated by a number
of vasodilatory metabolites produced by the neuron (Vasoactive Intestinal
Peptide, Nitric Oxide etc) which serve to increase regional CBF by acting on
local small vessels as well as upstream resistance vessels.41,42 The coupling
mechanism between neuronal activity and cerebral metabolism may be altered
in disease states such as cerebral ischaemia.

Ageing and cerebrovascular haemodynamics

Controversy exists regarding the effect of “normal ageing” on physiological
parameters such as CBF and cerebral metabolism. As with other aspects of
“normal ageing” and disease, the difficulty in establishing a clear relationship
between ageing and cerebrovascular physiology has been partly due to the
lack of a clear definition of “normal ageing”. Most studies examining this
relationship have been cross-sectional in nature (rather than longitudinal)
comparing younger to older age groups. Cross-sectional approaches lead to
difficulty in differentiating between genuine age-related change and cohort
effects unrelated to age. A great deal of uncertainty still exists about the
actual mechanisms involved in the perceived changes in CBF and cerebral
metabolism.

Ageing and CBF

Studies of ageing and CBF have been mostly cross-sectional in nature. “Nor-
mal ageing” in most of these studies refers to subjects who are relatively free
of vascular disease or risk factors for vascular disease. The results of most of
these studies, utilizing different techniques, seem to point towards a reduction
in mean CBF values with increasing age. These reductions have not only been
demonstrated in older age groups and may begin in the third or fourth dec-
CEREBROVASCULAR SYSTEM AND THE AGEING BRAIN 161

Table 1. Cross-sectional studies of cerebral blood flow and perfusion in human age-
ing.

Investigators Year Technique

Reduction of CBF with Ageing

Kety44 1956 Nitrous oxide
Wang et al.46 1975 133 Xenon inhalation
Obrist48 1979 133 Xenon inhalation
De Koninck et al.50 1977 133 Xenon intracarotid
Meyer et al.52 1978 133 Xenon inhalation
Yamaguchi et al.54 1979 133 Xenon inhalation
Naritomi et al.56 1979 133 Xenon inhalation
Thomas et al.58 1979 133 Xenon intravenous
Yamamoto et al.60 1980 133 Xenon inhalation
Melamed et al.62 1980 133 Xenon inhalation
Davis et al.64 1983 133 Xenon inhalation
Iwata et al.66 1986 131 Xenon CT
Zemcov et al.67 1984 133 Xenon inhalation
Gur et al.23 1987 133 Xenon inhalation
Hagstadius et al.68 1989 133 Xenon inhalation
Leenders et al.69 1990 P.E.T.
Martin et al.70 1991 P.E.T.
Markus et al.71 1993 99mTc-HMPAO SPECT
Krausz et al.72 1998 99mTc-HMPAO SPECT
Bentourkia et al.73 2000 P.E.T.
Scheel et al.74 2000 Colour duplex sonography

No Reduction of CBF with Ageing

Shieve and Wilson45 1953 Nitrous oxide
Shenkin et al.47 1953 Nitrous oxide
Gordan et al.49 1956 Nitrous oxide
Lassen et al.51 1960 85Kr
Dastur et al.53 1963 Nitrous oxide
Sokoloff et al.55 1966 Nitrous oxide
Aizawa et al.57 1961 Nitrous oxide
Gottstein et al.59 1979 Nitrous oxide
Waldemar61 1991 99mTc-HMPAO SPECT
Takada et al.63 1992 P.E.T.
Meltzer et al.65 2000 P.E.T.

P.E.T. – positron emission tomography; 99mTc-HMPAO SPECT – 99mTechnetium-

hexamethylpropyleneamine oxime single photon emission computed tomography.

ade of life. However there are a significant number of cross-sectional studies

with contradictory reports, with some evidence to support either a lack of or
a non-significant decline in CBF with increasing age (Table 1). This contro-
versy probably highlights the difficulty with inherent selection bias in using
cross-sectional designs, different subject populations and differing techniques
of measurement of CBF. In studies where CBF reduction with age has been
demonstrated, the reductions are more likely to involve cerebral grey matter
162 THE AGEING BRAIN

with an annual estimated decline of approximately 0.5 ml/min/100 g of brain

tissue. Evidence from current literature does not seem to support reduction
in white matter flow with healthy ageing.
Few longitudinal data exist with regards to CBF in ageing, presumably due
to the complexity and labour-intensive nature of CBF studies. A limited analy-
sis of CBF data in eight subjects over a 11-year period showed some reduction
in CBF in the absence of significant change in mean arterial blood pressure,
with a concomitant increase in cerebrovascular resistance.43 In a large cohort
of healthy volunteers, volunteers with vascular risks, and patients with stroke
or TIA, both cross-sectional and longitudinal analyses revealed significant
reduction in gray matter flow with age in all groups, with more decline dem-
onstrated in groups with risk factors or stroke/TIA.24 The estimated decline
in cerebral gray matter in this study was slightly higher than in cross-sectional
studies, estimated at approximately 1.0 ml/min/100 g of brain tissue.

Ageing and cerebral metabolism

Data regarding changes in the rate of cerebral metabolism with human ageing
are even less conclusive. Most of these data have been derived from positron
emission tomography studies (PET) measuring cerebral metabolic rate for
oxygen (CMRO2) and cerebral metabolic rate of glucose (CMRG) in resting
state (Table 2). Some investigators hold that CMRO2 declines in healthy age-
ing in parallel with decline in CBF. However a number of researchers have
failed to demonstrate reduction in mean CMRO2 levels with age. Similar
controversy exists for CMRG, with contradictory reports in the literature
regarding changes in glucose metabolism in “normal” ageing. The physiologi-
cal state of ‘coupling’ of cerebral glucose metabolism and CBF is thought to
be largely maintained with increasing age in the absence of disease.

Ageing, cerebrovascular reactivity and autoregulation

Cerebrovascular reactivity in response to changes in systemic arterial pressure
or to other parameters such as arterial CO2 and arterial O2 levels has been
investigated in only a few studies. Researchers examining CBF change with
posture showed some increase in the “dysautoregulation index” (defined as
CBF decrease per unit fall in effective perfusion pressure on head-up tilt). 88,89
Others have demonstrated a decline in CO2 reactivity with ageing using quanti-
tative CBF techniques.45,54,60 However, other groups using either quantitative
CBF techniques64 or transcranial doppler methods90–92 have failed to demon-
strate significant change in cerebral autoregulation with healthy ageing.

Ageing, disease and cerebrovascular haemodynamics

The more “common” form of ageing is one that is accompanied by the devel-
opment of disease as compared to the less common “healthy” (supranormal)
form of ageing. Studies of cerebrovascular haemodynamics have mostly been
performed in the latter group in order to gain insight into “normal”processes.
However, it is the study of the former group that is more relevant to the
CEREBROVASCULAR SYSTEM AND THE AGEING BRAIN 163

Table 2. Cross-sectional studies of cerebral metabolism in human ageing.

Investigator Year Method Comment

Reduced Cerebral Metabolism with Ageing

Sokoloff et al.75 1975 Nitrous oxide Reduction in CMRO2, CMRG and CBF
Kuhl et al.76 1984 18FDG PET Reduction in whole brain mean CMRG
Pantano et al.77 1984 H215O_PET Non-linear reduction in mean gray CBF
and CMRO2; white matter unchanged
Dastur78 1985 18FDG PET Reduction in CMRG; no reduction in CBF
and CMRO2
Yamaguchi et al.54 1986 H215O_PET Reduction in CMRO2; CBF variable and
less age-dependent
Leenders et al.69 1990 H215O_PET Coupled reduction in CMRO2 and CBF in
pure gray and white matter
Marchal et al.79 1992 H215O_PET Reduction in CMRO2 and CBV in gray
matter; CBF variable
Takada et al.63 1992 H215O_PET Reduction in CMRO2 but not CBF
De Santi et al.80 1995 18FDG PET Reduction in CMRG in frontal and tem-
poral lobes
Eberling et al.81 1995 18FDG PET Reduction in CMRG in temporal cortex
Bentourkia et al.73 2000 18FDG PET Coupled reduction in CBF and CMRG

Cerebral Metabolism Unchanged with Ageing

Duara et al.82 1984 18FDG PET No reduction in mean or regional CMRG;

resting CMRG poorly correlated with
cognitive tests
Cutler et al.83 1986 18FDG PET No reduction in mean CMRG
De Leon et al.84 1987 11CDG PET No reduction in absolute regional
CMRG
Horwitz et al.85 1987 18FDG PET No reduction in mean CMRG; age associ-
ated reduction in regional cerebral func-
tional interaction
Schlageter et al.86 1987 18FDG PET No reduction of global CMRG
Burns et al.87 1992 H215O_PET Trend for reduction in CMRO2 only in
parietal lobe; effect less significant with
advancing age

CMRO2: cerebral metabolic rate for oxygen; PET: positron emission tomogra-
phy; 18FDG: 18-Fluoro-deoxyglucose; CMRG: cerebral metabolic rate for glucose;
H215O: 15-oxygen labelled water; CBF: cerebral blood flow; 11CDG: 11-carbon-
deoxyglucose; CBV: cerebral blood volume.

majority of the ageing population and provides the opportunity to examine

the link between pathological vascular changes and a variety of disease states.
Certain vascular risk factors such as hypertension, diabetes, atrial fibrillation
and hypercholesterolaemia are common in the elderly population. The inci-
dence of disease states such as stroke and dementia increases with age. The
questions then arise as to whether a causal link may exist between vascular
164 THE AGEING BRAIN

risk and disease, and whether age plays a part in determining the expression
of such disease. Dementia (especially Alzheimer’s dementia) is a prototype
disease of the elderly that is the subject of intense interest and study with
regards to putative vascular aetiologies.
Epidemiological evidence exists from population-based studies describing
associations of vascular risk factors with prevalent dementia.93,94 The results
of studies of cerebral blood flow and metabolism performed in people with
dementia have indicated reductions in regional cerebral metabolism with con-
comitant reduction in regional CBF.24,78,95–97 These observations have led to
the debate whether the reduction in cerebral perfusion (putatively as a result
of vascular risk) plays an aetiological role in the development of dementia,98
or whether CBF reduction is a consequence of neuronal damage and the
vascular disease merely an epiphenomenon.96 Age by itself remains a crucial
risk factor in the development of dementia.99 Neuronal loss, oxidative stress,
reduction in vascular reserve and impaired repair mechanisms may all play a
role in reducing the reserve of the ageing brain thus leaving it vulnerable to
injury and disease.

Summary

A wide variety of changes are seen in the human cerebrovascular system on a

structural and functional basis as one gets older. However, there is a signifi-
cant amount of disagreement about the true nature of age-related vascular
changes in the brain. This is partly due to the increasing complexity in defin-
ing the boundaries of ‘normal’ ageing. The limits of human ageing are being
pushed further with every passing decade of medical scientific progress. As
people get to live longer, they become exposed to a larger number of vascu-
lar risks, predisposing them to accrue more vascular change. A more fruit-
ful approach towards unravelling this problem may be to identify the links
between such vascular risks, ageing and disease. This is likely to become a
reality with the increasingly sophisticated research methods now available.
The ultimate goal of such research is to develop ways to prevent and treat
important age-related neurological disorders.

References

1. Bouissou H, Emery MC, Sorbara R. Age related changes of the middle cerebral
artery and a comparison with the radial and coronary artery. Angiology. 1975;
26(3):257–68.
2. Fang HCH. Observations on aging characteristics of cerebral blood vessels, mac-
roscopic and microscopic features. In: Terry RD, Gershon, S., editors. Neurobiol-
ogy of aging. New York: Raven Press, 1976; 155–166.
3. Spangler KM, Challa VR, Moody DM, Bell MA. Arteriolar tortuosity of the white
matter in aging and hypertension. A microradiographic study. J Neuropath Exp
Neurol. 1994; 53:22–26.
CEREBROVASCULAR SYSTEM AND THE AGEING BRAIN 165

4. Mann DM, Eaves NR, Marcyniuk B, Yates PO. Quantitative changes in cerebral
cortical microvasculature in ageing and dementia. Neurobiol Aging. 1986; 7:
321–330.
5. Donahue J, Pappas G.D. The fine structure of capillaries in the cerebral cortex of
the rat at various stages of development. Am J Anat. 1961; 108:331–348.
6. Mooradian AD. Effect of aging on the blood-brain barrier. Neurobiol Aging.
1988; 9:31–39.
7. Burns EM, Kruckeberg TW, Comerford LE, Buschmann MT. Thinning of capil-
lary walls and declining numbers of endothelial mitochondria in the cerebral cor-
tex of the aging primate, Macaca nemestrina. J Gerontol. 1979; 34:642–650.
8. Stewart PA, Magliocco M, Hayakawa K, Farrell CL, Del Maestro RF, Girvin J,
Kaufmann JC, Vinters HV, Gilbert J. A quantitative analysis of blood-brain bar-
rier ultrastructure in the aging human. Microvasc Res. 1987; 33:270–282.
9. Moody DM, Brown WR, Challa VR, Ghazi-Birry HS, Reboussin DM. Cerebral
microvascular alterations in aging, leukoaraiosis, and Alzheimer’s disease. Ann
NY Acad Sci. 1997; 826:103–116.
10. Hardy JA, Mann DM, Wester P, Winblad B. An integrative hypothesis concerning
the pathogenesis and progression of Alzheimer’s disease. Neurobiol Aging. 1986;
7:489–502.
11. Wisniewski HM, Vorbrodt AW, Wegiel J. Amyloid angiopathy and blood-brain
barrier changes in Alzheimer’s disease. Ann NY Acad Sci. 1997; 826:161–172.
12. Alafuzoff I, Adolfsson R, Grundke-Iqbal I, Winblad B. Blood-brain barrier in
Alzheimer dementia and in non-demented elderly. An immunocytochemical study.
Acta Neuropathol. 1987; 73:160–166.
13. Kalaria RN. The blood-brain barrier and cerebrovascular pathology in Alzheim-
er’s disease. Ann NY Acad Sci .1999; 893:113–125.
14. Halliday G, Robinson SR, Shepherd C, Kril J. Alzheimer’s disease and inflam-
mation: a review of cellular and therapeutic mechanisms. Clin Exp Pharmacol P.
2000; 27:1–8.
15. Rozemuller JM, Eikelenboom P, Kamphorst W, Stam FC. Lack of evidence for
dysfunction of the blood-brain barrier in Alzheimer’s disease: an immunohisto-
chemical study. Neurobiol Aging. 1988; 9:383–391.
16. Munoz DG, Erkinjuntti T, Gaytan-Garcia S, Hachinski V. Serum protein leakage
in Alzheimer’s disease revisited. Ann NY Acad Sci. 1997; 826:173–189.
17. Vinters HV, Natte R, Maat-Schieman ML, van Duinen SG, Hegeman-Kleinn I,
Welling-Graafland C, Haan J, Roos RA. Secondary microvascular degeneration
in amyloid angiopathy of patients with hereditary cerebral hemorrhage with
amyloidosis, Dutch type (HCHWA- D). Acta Neuropathol. 1998; 95:235–244.
18. Garde E, Mortensen EL, Krabbe K, Rostrup E, Larsson HB. Relation between
age-related decline in intelligence and cerebral white- matter hyperintensities in
healthy octogenarians: a longitudinal study. Lancet. 2000; 356(9230):628–634.
19. Pantoni L, Garcia JH. Cognitive impairment and cellular/vascular changes in the
cerebral white matter. Ann NY Acad Sci. 1997; 826:92–102.
20. Moody DM, Brown WR, Challa VR, Anderson RL. Periventricular venous col-
lagenosis: association with leukoaraiosis. Radiology. 1995; 194:469–476.
21. Bronge L, Wahlund LO. White matter lesions in dementia: an MRI study on
blood-brain barrier dysfunction. Dement Geriatr Cogn. 2000; 11:263–267.
22. Werring DJ, Brassat D, Droogan AG, Clark CA, Symms MR, Barker GJ, Mac-
Manus DG, Thompson AJ, Miller DH. The pathogenesis of lesions and normal-
appearing white matter changes in multiple sclerosis: a serial diffusion MRI study.
Brain. 2000; 123:1667–1676.
23. Gur RC, Gur RE, Obrist WD, Skolnick BE, Reivich M. Age and regional cerebral
blood flow at rest and during cognitive activity. Arch Gen Psychiat. 1987; 44:
617–621.
166 THE AGEING BRAIN

24. Shaw TG, Mortel KF, Meyer JS, Rogers RL, Hardenberg J, Cutaia MM. Cer-
ebral blood flow changes in benign aging and cerebrovascular disease. Neurology.
1984; 34:855–862.
25. Thomas DJ. Whole blood viscosity and cerebral blood flow. Stroke. 1982; 13:
285–287.
26. Humphrey PR, Du Boulay GH, Marshall J, Pearson TC, Russell RW, Symon L,
Wetherley-Mein G, Zilkha E. Cerebral blood-flow and viscosity in relative poly-
cythaemia. Lancet. 1979; 2(8148):873–877.
27. Paulson OB, Strandgaard S, Edvinsson L. Cerebral autoregulation. Cerebrovas
Brain Met. 1990; 2:161–192.
28. Edvinsson L, Owman C, Siesjo B. Physiological role of cerebrovascular sympa-
thetic nerves in the autoregulation of cerebral blood flow. Brain Res. 1976; 117:
519–523.
29. Edvinsson L, Degueurce A, Duverger D, MacKenzie ET, Scatton B. Central serotoner-
gic nerves project to the pial vessels of the brain. Nature. 1983; 306(5938):55–57.
30. Hossmann KA. Treatment of experimental cerebral ischemia. J Cerebr Blood
Met. 1982; 2:275–297.
31. Bayliss WM. On the local reaction of the arterial wall to changes of internal pres-
sure. J Physiol. 1902:220–231.
32. Harder DR. Pressure-induced myogenic activation of cat cerebral arteries is
dependent on intact endothelium. Circ Res. 1987; 60:102–107.
33. Harder DR, Sanchez-Ferrer C, Kauser K, Stekiel WJ, Rubanyi GM. Pressure
releases a transferable endothelial contractile factor in cat cerebral arteries. Circ
Res. 1989; 65:193–198.
34. Roy CS, Sherrington, C.S. On the regulation of the blood supply of the brain. J
Physiol .1890; 11:85–108.
35. Swanson RA, Morton MM, Sagar SM, Sharp FR. Sensory stimulation induces
local cerebral glycogenolysis: demonstration by autoradiography. Neuroscience.
1992; 51:451-461.
36. Jueptner M, Weiller C. Review: does measurement of regional cerebral blood flow
reflect synaptic activity? Implications for PET and fMRI. Neuroimage. 1995; 2:
148–156.
37. Sibson NR, Dhankhar A, Mason GF, Rothman DL, Behar KL, Shulman RG.
Stoichiometric coupling of brain glucose metabolism and glutamatergic neuronal
activity. Proc Natl Acad Sci USA. 1998; 95:316-321.
38. Gerrits RJ, Raczynski C, Greene AS, Stein EA. Regional cerebral blood flow
responses to variable frequency whisker stimulation: an autoradiographic analy-
sis. Brain Res. 2000; 864:205–212.
39. Baron JC, Lebrun-Grandie P, Collard P, Crouzel C, Mestelan G, Bousser MG.
Noninvasive measurement of blood flow, oxygen consumption, and glucose utili-
zation in the same brain regions in man by positron emission tomography: concise
communication. J Nucl Med. 1982; 23:391–399.
40. Baron JC, Rougemont D, Soussaline F, Bustany P, Crouzel C, Bousser MG,
Comar D. Local interrelationships of cerebral oxygen consumption and glucose
utilization in normal subjects and in ischemic stroke patients: a positron tomog-
raphy study. J Cerebr Blood F Met. 1984; 4:140–149.
41. Akgoren N, Dalgaard P, Lauritzen M. Cerebral blood flow increases evoked by
electrical stimulation of rat cerebellar cortex: relation to excitatory synaptic activ-
ity and nitric oxide synthesis. Brain Res. 1996; 710:204–214.
42. Ngai AC, Ko KR, Morii S, Winn HR. Effect of sciatic nerve stimulation on pial
arterioles in rats. Am J Physiol. 1988; 254:H133–139.
43. Libow LS. Cerebral and electroencephalographic changes in elderly men. Rock-
ville, US Dep. of Health, Education and Welfare, National Institute of Mental
Health, 1971. Report No: (HSM) 71–9037.
CEREBROVASCULAR SYSTEM AND THE AGEING BRAIN 167

44. Kety SS. Human cerebral blood flow and oxygen consumption as related to aging.
Res Publ Assoc Res N. 1956; 35:31–35.
45. Schieve JF, Wilson, WP. The influence of age, anesthesia and cerebral arterioscle-
rosis on cerebral vascular reactivity to CO2. Am J Med. 1953; 15:171–174.
46. Wang HS, Busse EW. Correlates of regional blood flow in elderly community
residents. In: Harper M, Jennett, B, Miller, D, Rowan, J, editors. Blood flow and
metabolism in the brain. London: Churchill Livingstone, 1975; 17–18.
47. Shenkin HA, Novak, P., Golobuff, B., Soffe, A.M., Bortin, L. The effects of aging
arteriosclerosis, and hypertension upon the cerebral circulation. J Clin Invest.
1953; 32:459–465.
48. Obrist WD. Cerebral circulatory changes in normal aging and dementia. In: Bayer
Symposium VII;Brain function in old age. New York: Springer-Verlag, 1979;
278–287.
49. Gordan GS. Influence of steroids on cerebral metabolism in man. Recent Prog
Horm Res. 1956; 12:153–174.
50. De Koninck WJ, Calay, R., Hongne, JC. CBF in elderly with chronic cerebral
involvement. Acta Neurol Scand. 1977; (Suppl) 64:412–413.
51. Lassen NA, Feinberg, I., Lane, M.H. Bilateral studies of cerebral oxygen uptake
in young and aged normal subjects and in patients with organic dementia. J Clin
Invest. 1960; 39:491–500.
52. Meyer JS, Ishihara N, Deshmukh VD, Naritomi H, Sakai F, Hsu MC, Pollack
P. Improved method for noninvasive measurement of regional cerebral blood
flow by 133Xenon inhalation. Part I: description of method and normal values
obtained in healthy volunteers. Stroke. 1978; 9:195–205.
53. Dastur DK, Lane MH, Hansen, DB. Effects of aging on cerebral circulation and
metabolism in man. Washington DC: US Government Printing Office, 1963.
USPHS publication no. 986.
54. Yamaguchi T, Kanno I, Uemura K, Shishido F, Inugami A, Ogawa T, Murakami
M, Suzuki K. Reduction in regional cerebral metabolic rate of oxygen during
human aging. Stroke. 1986; 17:1220–1228.
55. Sokoloff L. Cerebral circulatory and metabolic changes associated with aging.
Res Publ Assoc Res N. 1966; 41:237–254.
56. Naritomi H, Meyer JS, Sakai F, Yamaguchi F, Shaw T. Effects of advancing age
on regional cerebral blood flow. Studies in normal subjects and subjects with risk
factors for atherothrombotic stroke. Arch Neurol. 1979; 36:410–416.
57. Aizawa T, Tazaki, Y., Gotoh, F. Cerebral circulation in cerebrovascular disease.
World Neurol. 1961; 2:635–45.
58. Thomas DJ, Zilkha E, Redmond S, Du Boulay GH, Marshall J, Russell RW,
Symon L. An intravenous 133xenon clearance technique for measuring cerebral
blood flow. J Neurol Sci. 1979; 40:53–63.
59. Gottstein U, Held, K. Effects of aging on cerebral circulation and metabolism in
man. Acta Neurol Scand. 1979; (Suppl) 72:54–55.
60. Yamamoto M, Meyer JS, Sakai F, Yamaguchi F. Aging and cerebral vasodilator
responses to hypercarbia: responses in normal aging and in persons with risk fac-
tors for stroke. Arch Neurol. 1980; 37:489–496.
61. Waldemar G, Hasselbalch SG, Andersen AR, Delecluse F, Petersen P, Johnsen
A, Paulson OB. 99mTc-d,l-HMPAO and SPECT of the brain in normal aging. J
Cerebr Blood F Met. 1991; 11:508–521.
62. Melamed E, Lavy S, Bentin S, Cooper G, Rinot Y. Reduction in regional cerebral
blood flow during normal aging in man. Stroke. 1980; 11:31–35.
63. Takada H, Nagata K, Hirata Y, Satoh Y, Watahiki Y, Sugawara J, Yokoyama
E, Kondoh Y, Shishido F, Inugami A. Age-related decline of cerebral oxygen
metabolism in normal population detected with positron emission tomography.
Neurol Res. 1992; 14:128–131.
168 THE AGEING BRAIN

64. Davis SM, Ackerman RH, Correia JA, Alpert NM, Chang J, Buonanno F, Kelley
RE, Rosner B, Taveras JM. Cerebral blood flow and cerebrovascular CO2 reactiv-
ity in stroke-age normal controls. Neurology. 1983; 33:391–399.
65. Meltzer CC, Cantwell MN, Greer PJ, Ben-Eliezer D, Smith G, Frank G, Kaye WH,
Houck PR, Price JC. Does cerebral blood flow decline in healthy aging? A PET
study with partial-volume correction. J Nucl Med. 2000; 41:1842–1848.
66. Iwata K, Harano H. Regional cerebral blood flow changes in aging. Acta Radiol
Suppl. 1986; 369:440–443.
67. Zemcov A, Barclay L, Blass JP. Regional decline of cerebral blood flow with age
in cognitively intact subjects. Neurobiol Aging. 1984; 5:1–6.
68. Hagstadius S, Risberg J. Regional cerebral blood flow characteristics and varia-
tions with age in resting normal subjects. Brain Cogn. 1989; 10:28–43.
69. Leenders KL, Perani D, Lammertsma AA, Heather JD, Buckingham P, Healy MJ,
Gibbs JM, Wise RJ, Hatazawa J, Herold S. Cerebral blood flow, blood volume and
oxygen utilization. Normal values and effect of age. Brain. 1990; 113:27–47.
70. Martin AJ, Friston KJ, Colebatch JG, Frackowiak RS. Decreases in regional cer-
ebral blood flow with normal aging. J Cerebr Blood F Met. 1991; 11:684–689.
71. Markus HS, Ring H, Kouris K, Costa DC. Alterations in regional cerebral blood
flow, with increased temporal interhemispheric asymmetries, in the normal eld-
erly: an HMPAO SPECT study. Nucl Med Commun. 1993; 14:628–633.
72. Krausz Y, Bonne O, Gorfine M, Karger H, Lerer B, Chisin R. Age-related changes
in brain perfusion of normal subjects detected by 99mTc-HMPAO SPECT. Neu-
roradiology. 1998; 40):428–434.
73. Bentourkia M, Bol A, Ivanoiu A, Labar D, Sibomana M, Coppens A, Michel
C, Cosnard G, De Volder AG. Comparison of regional cerebral blood flow and
glucose metabolism in the normal brain: effect of aging. J Neurol Sci. 2000; 181:
19–28.
74. Scheel P, Ruge C, Petruch UR, Schoning M. Color duplex measurement of cer-
ebral blood flow volume in healthy adults. Stroke. 2000; 31:147–150.
75. Sokoloff L. Cerebral circulation and metabolism in the aged. Psychopharmacol
Bull. 1975; 11:45-46.
76. Kuhl DE, Metter EJ, Riege WH, Hawkins RA. The effect of normal aging on
patterns of local cerebral glucose utilization. Ann Neurol. 1984; 15(Suppl):
S133–137.
77. Pantano P, Baron JC, Lebrun-Grandie P, Duquesnoy N, Bousser MG, Comar D.
Regional cerebral blood flow and oxygen consumption in human aging. Stroke.
1984; 15:635–641.
78. Dastur DK. Cerebral blood flow and metabolism in normal human aging, patho-
logical aging, and senile dementia. J Cerebr Blood F Met. 1985; 5:1–9.
79. Marchal G, Rioux P, Petit-Taboue MC, Sette G, Travere JM, Le Poec C, Courthe-
oux P, Derlon JM, Baron JC. Regional cerebral oxygen consumption, blood flow,
and blood volume in healthy human aging. Arch Neurol. 1992; 49:1013–1020.
80. De Santi S, de Leon MJ, Convit A, Tarshish C, Rusinek H, Tsui WH, Sinaiko E,
Wang GJ, Bartlet E, Volkow N. Age-related changes in brain: II. Positron emis-
sion tomography of frontal and temporal lobe glucose metabolism in normal
subjects. Psychiatr Quart. 1995; 66:357–370.
81. Eberling JL, Nordahl TE, Kusubov N, Reed BR, Budinger TF, Jagust WJ. Reduced
temporal lobe glucose metabolism in aging. J Neuroimaging. 1995; 5:178–182.
82. Duara R, Grady C, Haxby J, Ingvar D, Sokoloff L, Margolin RA, Manning RG,
Cutler NR, Rapoport SI. Human brain glucose utilization and cognitive function
in relation to age. Ann Neurol. 1984; 16:703–713.
83. Cutler NR. Cerebral metabolism as measured with positron emission tomography
(PET) and [18F] 2-deoxy-D-glucose: healthy aging, Alzheimer’s disease and Down
syndrome. Prog Neuro-psychoph. 1986; 10:309–321.
CEREBROVASCULAR SYSTEM AND THE AGEING BRAIN 169

84. de Leon MJ, George AE, Tomanelli J, Christman D, Kluger A, Miller J, Ferris
SH, Fowler J, Brodie JD, van Gelder P. Positron emission tomography studies of
normal aging: a replication of PET III and 18-FDG using PET VI and 11-CDG.
Neurobiol Aging. 1987; 8:319–323.
85. Horwitz B. Brain metabolism and blood flow during aging. Electroen Clin Neuro
Suppl. 1987; 39:396–402.
86. Schlageter NL, Horwitz B, Creasey H, Carson R, Duara R, Berg GW, Rapoport
SI. Relation of measured brain glucose utilisation and cerebral atrophy in man. J
Neurol Neurosur Ps. 1987; 50:779-785.
87. Burns A, Tyrrell P. Association of age with regional cerebral oxygen utilization:
a positron emission tomography study. Age Ageing. 1992; 21:316–320.
88. Shinohara Y, Takagi S, Kobatake K. Effect of aging on CBF and autoregulation
in normal subjects and CVD patients. Monogr Neural Sci. 1984; 11:204–209.
89. Oiwa K, Shimazu K, Tamura N, Hienuki M, Kim HT, Yamamoto T, Hamaguchi
K. Effect of aging on cerebral blood flow autoregulation--with special reference
to the role of the prostaglandins. Monogr Neural Sci. 1984; 11:210–215.
90. Kastrup A, Dichgans J, Niemeier M, Schabet M. Changes of cerebrovascular CO2
reactivity during normal aging. Stroke. 1998; 29:1311–1314.
91. Carey BJ, Eames PJ, Blake MJ, Panerai RB, Potter JF. Dynamic cerebral autoregu-
lation is unaffected by aging. Stroke. 2000; 31(12):2895–2900.
92. Lipsitz LA, Mukai S, Hamner J, Gagnon M, Babikian V. Dynamic regulation
of middle cerebral artery blood flow velocity in aging and hypertension. Stroke.
2000; 31(8):1897–1903.
93. Hofman A, Ott A, Breteler MM, Bots ML, Slooter AJ, van Harskamp F, van Duijn
CN, Van Broeckhoven C, Grobbee DE. Atherosclerosis, apolipoprotein E, and
prevalence of dementia and Alzheimer’s disease in the Rotterdam Study. Lancet.
1997; 349(9046):151–154.
94. Elwood PC, Pickering J, Gallacher JE. Cognitive function and blood rheol-
ogy: results from the Caerphilly cohort of older men. Age Ageing. 2001; 30:
135–139.
95. Benson DF, Kuhl DE, Hawkins RA, Phelps ME, Cummings JL, Tsai SY. The
fluorodeoxyglucose 18F scan in Alzheimer’s disease and multi- infarct dementia.
Arch Neurol. 1983; 40:711–714.
96. Jagust WJ, Eberling JL, Reed BR, Mathis CA, Budinger TF. Clinical studies of cer-
ebral blood flow in Alzheimer’s disease. Ann NY Acad Sci. 1997; 826:254–262.
97. Tohgi H, Yonezawa H, Takahashi S, Sato N, Kato E, Kudo M, Hatano K, Sasaki
T. Cerebral blood flow and oxygen metabolism in senile dementia of Alzheimer’s
type and vascular dementia with deep white matter changes. Neuroradiology.
1998; 40:131–137.
98. de la Torre JC, Stefano GB. Evidence that Alzheimer’s disease is a microvas-
cular disorder: the role of constitutive nitric oxide. Brain Res Rev. 2000; 34:
119–136.
99. Munoz DG, Feldman H. Causes of Alzheimer’s disease. Cmaj. 2000; 162:
65–72.
 SECTION III

FACTORS
INFLUENCING
BRAIN AGEING
Chapter 10

THE MOLECULAR BASIS

OF ALZHEIMER’S DISEASE
AND FRONTOTEMPORAL
DEMENTIA
John B.J. Kwok and Peter R. Schofield*

Introduction

Alzheimer’s disease (AD) was thought to be an intractable disorder. Yet,

genetic analyses have successfully uncovered three genes, the amyloid precur-
sor protein (APP) gene, presenilin-1 (PS-1) and the presenilin-2 (PS-2) gene
which can cause early-onset Alzheimer’s disease. Functional analysis of these
genes and gene mutations has highlighted the importance of the amyloid
cascade hypothesis to our understanding of the disease process. Moreover,
the correlation of mutations in the AD genes with specific clinical outcomes
and variant neuropathology has allowed us to detect the existence of modi-
fying factors which alter the course of the disease. More recently, the tau
gene has been identified as the causative agent for another form of dementia,
fronto-temporal dementia. The challenge now is to determine the enigmatic
relationship between tau and the AD genes and to determine whether there
is a common neurodegenerative mechanism.

*To whom correspondence should be addressed.

174 THE AGEING BRAIN

A Common Affliction

Alzheimer’s disease (AD) is a devastating affliction of the brain. A patient

will suffer an irreversible deterioration of intellectual abilities involving
memory loss, impairment of judgement and reasoning, as well as personal-
ity changes in later stages. Ultimately, the condition is fatal due to failure
of physical function.1 AD is the most common cause of senile dementia,
accounting for 50% of dementia cases. The disease will strike an estimated
one in ten persons over the age of 65 years and increases to nearly one in
two of those over 80.2 Despite intense basic scientific and pharmacological
investigations, there are still no truly effective therapeutic drugs for AD.
With such an emotional and financial cost to patients, and to a rapidly
ageing society, there is a pressing need for greater understanding of this
disease.
AD is distinct from other forms of dementia by key pathological features
in the brain. Firstly, there are a large number of senile plaques in the extracel-
lular spaces between neurones. The plaques are spherical deposits that consist
of central cores of amyloid beta (Aβ) fibrils, surrounded by degenerated neu-
rites and glial cells.2 The Aβ peptide consists of a sequence of hydrophobic
amino acids of 39 to 43 amino acids in length.3,4 The peptide is derived
from proteolytic cleavage of a larger multidomain glycoprotein, the amyloid
precursor protein (APP).5 Secondly, there are neurofibrillary tangles (NFTs)
found within neurones. The major components of NFTs are paired helical
filaments, which in turn are composed of hyperphosphorylated form of tau,
a microtubule associated protein2. Finally, there is extensive neuronal loss in
the cerebral cortex and hippocampus, which is directly responsible for the
cognitive decline.2

Amyloid Cascade

The exact contribution of each neuropathological feature to the clinical

symptoms of AD is unclear.6 However, several studies have suggested that
the Aβ deposits can be directly neurotoxic, in part through the generation
of free radicals6 whose effects can be attenuated by the addition of antioxi-
dants.6 Other studies suggest that Aβ can disrupt ionic homeostasis and lead
to severe effects on cellular processes and induction of neuronal cell death6.
The amyloid cascade hypothesis postulates that the deposition of Aβ is the
central causative event in AD and that the NFTs, cell death and dementia
follow as a direct result of this deposition7 as shown in Figure 1. The amy-
loid cascade hypothesis predicts that mutations in genes, which lead to the
overproduction of APP or subsequent mismetabolism, would underlie the
genetic basis of AD.
ALZHEIMER’S DISEASE AND FRONTOTEMPORAL DEMENTIA 175

Figure 1. Amyloid Cascade hypothesis suggests the aberrant metabolism of APP mol-
ecule to form the longer peptide isoform, Aβ1-42 and greater deposition of
senile plaques. This may arise as a result of age-related factors or genetic
mutations. Neurofibrillary tangles and neuronal cell death are secondary
events to the initial amyloid production and deposition.

Amyloid Precursor Protein (APP) Gene

The Aβ peptide is cleaved from APP via a series of proteolytic steps mediated
by enzymes called secretases (Figure 2). Cleavage of APP with the β- and
γ-secretase will generate an intact Aβ peptide. Cleavage by the α-secretase
within the peptide sequence will prevent the formation of Aβ.5 All the muta-
tions appear to cluster within or adjacent to the sequence which encode Aβ
peptide as shown in Figure 2. Each mutation has an effect, either on the
metabolism of APP or the nature of the Aβ sequence itself, but ultimately all
mutations increase the rate of amyloid deposition.8 For example, the Swedish
double mutation results in increased secretion of the normal 40 amino-acid
peptide (Aβ1-40) and a longer 42 amino-acid isoform of Aβ (Aβ1-42), most
probably by enhancing β-secretase activity. The cerebral angiopathy with
amyloidosis (CAA) mutation has been shown to diminish α-secretase activ-
ity, thus increasing the secretion of both forms of intact Aβs. Finally, there
are a series of mutations which cluster around the γ-secretase site (Figure
2). These mutations include the London mutation at codon 717 (Val to Ile),
176 THE AGEING BRAIN

Figure 2. Schematic diagram of APP. The protein is a multi-domain cell-surface

molecule. The Aβ region (grey box and circles) contains part of the trans-
membrane domain and part of the extracellular domain of APP. Secretase
cleavage sites are indicated by open arrows. Familial mutations that cause
AD are indicated.

which alters the conformation of the γ-secretase recognition site so that APP
is preferentially cleaved to produce the Aβ1-42 isoform.9

Presenilin Genes

A major locus responsible for early onset AD (EOAD), presenilin-1 (PS-1),

was shown to map to the long arm of chromosome 14.10 Together with its
homologue, presenilin-2 (PS-2), on chromosome 1,11 pathogenic mutations in
ALZHEIMER’S DISEASE AND FRONTOTEMPORAL DEMENTIA 177

Figure 3. Schematic diagram of PS-1. The protein has eight hydrophobic trans-
membrane spanning domains and a large hydrophilic loop. The protein is
cleaved within the hydrophilic loop by an unknown protease (presenilinase)
and a caspase. The majority of mutations detected in EOAD pedigrees and
cases are missense mutations (mutant amino indicated in grey). One special
class of mutations which deletes exon 9 (boundaries indicated by open
arrows) is associated with variable neuropathology and differing clinical
presentations.

these genes account for approximately 50% of EOAD cases.12 As shown in

Figure 3, presenilin-1 is predicted to code for a novel transmembrane protein
with up to nine potential hydrophobic domains.5 Over 70 point mutations,
and splice-site mutations which results in the deletion of a small portion of
the protein, have been identified in the presenilin genes.13 These mutations
span all putative domains of the presenilin protein, including every potential
transmembrane domain as shown in Figure 3. Biochemical analyses indicate
all mutations result in the elevated secretion of the amyloidogenic Aβ1–42
isoform.14 Transgenic mice and clonal cell lines which express mutant forms
of the presenilin genes have elevated production of Aβ1–42 compared with
wildtype constructs.15 Moreover, the elevated secretion of Aβ1-42 peptide was
also found in plasma and fibroblasts cultured from AD patients with known
PS-1 mutations.16 The mechanism of presenilin-induced Aβ1-42 elevation is
still being debated.17 Several lines of evidence suggest that PS-1 is a diaspartyl
178 THE AGEING BRAIN

protease which serves as the γ-secretase in the processing of APP. Mutation

of two conserved aspartate residues at position 257 and 385 of PS-1, which
face the predicted catalytic site, abrogated γ-secretase actvity.18 Moreover,
both photoactivated and transition state analogues of γ-secretase inhibitors
appear to bind directly to PS-1.19,20

AD Variants — Cotton Wool Plaques and Spastic Paraparesis

The hallmark feature of AD is the presence of a large number of neuritic

plaques and neurofibrillary tangles (NFTs). Mutations in the PS-1 gene are
normally associated with severe neuropathology. Typically, there are large
numbers of diffuse as well as cored, neuritic plaques, which are deposited in
the cerebral cortex.21 Moreover, there is intense tau pathology, with neuritic
dystrophy and NFTs.22,23 However, variations exist in AD neuropathology
which include the morphology of senile plaques and the level of tau pathol-
ogy. Three mutations in PS-1, a deletion of exon 9 (PS-1Δexon9),24,25 a dou-
ble amino acid deletion (ΔI83/ΔM84),26 and a misssense mutation (P436Q)27
have been associated with a variant, “cotton wool” plaque pathology. As
shown in Figure 4, a brain from an affected individual carrying the (PS-
1Δexon9) mutations had extensive deposition of large, spherical plaques that
lacked distinctive cores and neuritic dystrophy. Biochemical analysis of cells
transfected with PS-1 cDNAs carrying either of the three mutations secrete
exceptionally high levels of Aβ1-4227 and this was suggested to be the molecu-
lar basis for PS-1 induced variant plaques.
The main clinical symptoms of AD are initial memory deficits and progres-
sive loss of higher cognitive function. However, variant forms of AD have
been reported which manifest other neurological disorders. Spastic parapare-
sis (SP), or progressive spasticity of the lower limbs, frequently occurs on a
hereditary background and a number of reports have described SP in families
with dementia.13,24,27,28 Mutations reported in familial AD with SP have
been confined to PS-1 with the majority of mutations consisting of deletions
of exon 9.25,27,29 In the Finn2 pedigree, which has a PS-1Δexon9 mutation,24
10 of 14 individuals with dementia also had SP. Examination of the brains
of three subjects, two with and one without SP, revealed many cotton wool
plaques in all three cases, together with NFTs and pronounced congophilic
angiopathy.24 This led to the suggestion there was an association between
the variant plaques and SP clinical presentation. Further investigations into
the pattern of inheritance of the dementia and SP phenotypes within another
branch of the Finnish pedigree,30 and in an Australian pedigree25 (Figure 4a),
suggests that the SP phenotype is due to the inheritance of an unlinked genetic
locus acting in concert with the PS-1 mutation.
A characteristic of the EOAD pedigrees with PS-1 mutations is that the
mutation is usually fully penetrant and that there is a narrow range of age
of onset of the disease within family members.31 However, there are excep-
ALZHEIMER’S DISEASE AND FRONTOTEMPORAL DEMENTIA 179

Figure 4. Spastic paraparesis and AD in an Australian EOAD pedigree. (A) Vari-

able presentation of clinical phenotypes. Left-half black symbols indicate
dementia without spastic paraparesis, right-half black symbols indicate
spastic paraparesis without apparant dementia and filled symbols indicates
spastic paraparesis with dementia. Haplotype analysis using four microsat-
ellite markers flanking the PS-1 gene reveals a common disease haplotype
(open box) detected in the affected individuals. (B) Variable neuropathol-
ogy in EOAD pedigree as detected by antibodies against Aβ (upper panel)
or tau (lower panel). Cotton wool plaques are distinguished by their lack
of a distinctive amyloid core and neuritic dystrophy which were mostly
detected in the individual with SP (III:9).

tions. A pedigree with the H163T mutation has been reported which has a
non-penetrant individual as well as a wide range of age of onset.32 This sup-
ports the existence of genes and/or environmental factors that may modulate
the expression of the AD phenotype. Analysis of large numbers of sib pairs
affected with AD33 and ascertainment of the risk of relatives of aged, non-
demented probands to develop AD,34 have indicated that there are genetic
factors which are protective against AD or modify the age of onset of the
disease. One such factor is the apolipoprotein E (ApoE) gene on chromosome
19. Inheritance of the ApoE ε4 allele has been shown to increase the risk of
late-onset AD (LOAD) and results in an earlier age of onset in EOAD pedi-
grees with heritable mutations in the APP gene.35 Conversely, inheritance of
the ApoE ε2 allele delays the onset of EOAD and LOAD cases.36 Another pos-
180 THE AGEING BRAIN

sible factor is the butyrylcholinesterase (BChE) gene on chromosome 3, which

may be protective against LOAD.37 The clinical phenotype in our subjects
with SP is unusual in that dementia onset appears to be delayed compared to
affected individuals who presented with dementia only, since three of the four
individuals who developed SP remained dementia-free for up to ten years.25
Thus, the study of variant forms of AD has been important in suggesting the
existence of phenotypic modifier gene(s) act in concert with specific PS-1
mutations to alter the clinical presentation of the disease.

Frontotemporal Dementia and the tau Gene

Frontotemporal dementias (FTD) represent a significant group of degenera-

tive disorders that overlap in their clinical and neuropathological descrip-
tions. These include seemingly separate disorders such as pallido-ponto-nigral

Figure 5. Mutations in tau detected in FTDP-17. The largest tau isoform is shown
within residues from alternatively spliced exons 2,3 and 10. Gray boxes
represent each of the four microtubule binding domains. The stem loop
structure of the 5’ splice donor site of exon 10 is drawn above the tau
isoform (not to scale). Exonic sequence is given in upper case and intronic
sequence is in lower case letters.
ALZHEIMER’S DISEASE AND FRONTOTEMPORAL DEMENTIA 181

Figure 6. Exon trapping analysis of tau mutations. (A) Schematic diagram showing
the exon trap procedure. Genomic DNA containing exon 10 of the tau gene
is subcloned into the pSPL3 exon trap vector between two vector splice
sites. The recombinant construct is transfected into cells whereby the vector
promoter drives expression of chimaric RNA. In vivo splicing will generate
either mRNA which includes or excludes exon 10 sequence. The exon trap
products are then detected by RT-PCR. (B) Gel electrophoresis of RT-PCR
products from exon trapping assays in COS-7 cells. The splicing in of exon
10 yeilds a 270 bp product, while the absence results in a 177bp product.
The normal tau exon 10 (wt) shows the presence of both exon 10-positive
(E10+) and exon 10-negative (E10-) RT-PCR products. The +16 and S305S
mutations result in a marked increase in E10+ products. The +19 and 29
mutations result in a marked decrease in E10+ products.
182 THE AGEING BRAIN

degeneration, familial multiple system tauopathy, familial progressive sub-

cortical gliosis, progressive supranuclear palsy and disinhibition-dementia-
parkinsonism-amyotrophy complex. However, as a group, these disorders
share many clinical and neuropathological similarities and the tau gene has
been genetically linked to a number of such pedigrees which have now been
defined as Frontotemporal Dementia with Parkinsonism (FTDP-17).38
Tau is a microtubule-associated protein that is involved in the neuronal
cytoskeleton, in particular, the assembly and stability of microtubules. Alter-
native splicing of exon 10 in the carboxy-terminal half of the protein result
in tau protein which contains either three or four of the microtubule-binding
repeat motifs (3 repeat tau and 4 repeat tau respectively). In 1998, tau gene
mutations were shown to be causal for FTDP-17.39–41 Tau mutations can be
functionally divided into two groups as shown in Figure 5. Firstly, missense
mutations such as P301L and V337M have been shown to result in mutant
forms of tau with decreased affinity for binding to mictotubules. The second
group of exonic or intronic mutations alter the efficiency of splicing of exon
10.42 Intronic mutations (+3, +11, +12, +13, +14 and +16) identified in the
5’ splice donor site of intron 10 cause an increase in splicing of exon 10 by
disrupting a stem-loop structure (Figure 5). The effect of a mutation on splic-
ing can be assayed using the in vitro exon trapping assay which analyses the
ability of the splice sites flanking an exon of interest to be spliced onto an
exon trap vector’s reporter splice donor and reporter site43 (Figure 6A). As
shown in Figure 6B, the presence of the +16 mutation results in an increase
in exon trap products which contains the spliced exon 10 sequence compared
with wildtype sequence.

A Common Biochemical Pathway?

One of the most crucial questions in AD research is the enigmatic relation-

ship between senile plaques and NFTs. Studies have shown a direct causal
relationship between the two diagnostic features of AD, in which Aβ appears
to induce the hyperphosphorylation of tau. For example, when transgenic
mice, which overexpress mutant tau (Pro301Leu) are injected with Aβ1-42
fibrils, NFTs are induced.44 Moreover, there are mutations in the tau gene
that causes tau polymerisation and NFTs, but are associated with FTD rather
than AD. Conversely, all mutations in the AD genes give rise to the develop-
ment of tau pathology. These studies support the amyloid cascade hypothesis
which states that the production and deposition of Aβ is the primary event in
AD aetiology, which then cause the secondary formation of NFTs.
There has been a spirited debate within the AD research field as to the
relevance of Aβ1-42 and the amyloid cascade hypothesis to AD aetiology. Yet,
it is still possible that there is a more fundamental neurodegenerative mecha-
nism at work. There have been tantalising studies which suggest that there is a
common cytotoxic mechanism involved in both AD and FTD. A recent study
ALZHEIMER’S DISEASE AND FRONTOTEMPORAL DEMENTIA 183

has shown that the apoptotic pathway involving the caspases is activated in
AD brains.45 In vitro studies using cells transfected with mutant presenilin
genes or exposed to Aβ peptides demonstrated that the cells undergo apop-
tosis.46,47 Similar experiments have shown that cells transfected with mutant
tau cDNAs undergo the same process.48 It is hoped that the insights into
the fundamental pathological mechanisms of the AD and FTD genes can be
applied to the design of an effective prophylactic or therapeutic strategy for
all forms of dementia. Knowing the specific molecules involved in this proc-
ess, which has been achieved through genetic studies, is a crucial first step in
the process.

Acknowledgements

Supported by a Department of Veterans Affairs Research Grant 9937441 and

the National Health and Medical Research Council Block Grant and Research
Fellowship 993050 and Unit Grant 983302.

References

1. Iqbal K, ed. Research advances in Alzheimer’s Disease and related disorders.

Chicester: John Wiley and Sons, 1995.
2. Clark RF, Goate AM. Molecular genetics of Alzheimer’s disease. Arch Neurol.
1993; 50:1164–1172.
3. Masters CL, Simms G, Weinman NA, Multhaup G, McDonald BL, Beyreuther
K. Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc
Natl Acad Sci USA. 1985; 82:4245–4249.
4. Selkoe DJ, Abraham CR, Podlisny MB, Duffy LK. Isolation of low-molecular-
weight proteins from amyloid plaque fibers in Alzheimer’s disease. J Neurochem.
1986; 146:1820–1834.
5. Price DL, Sisodia SS. Mutant genes in familial Alzheimer’s disease and transgenic
models. Annu Rev Neurosci. 1998; 21:479–505.
6. Drouet B, Pincon-Raymond M, Chambaz J, Pillot T. Molecular basis of Alzheim-
er’s disease. Cell Mol Life Sci. 2000; 57:705–715.
7. Hardy JA, Higgins GA. Alzheimer’s disease: the amyloid cascade hypothesis.
Science. 1992; 256:184–185.
8. Hardy J. Amyloid, the presenilins and Alzheimer’s disease. Trends Neurosci.
1997; 20:154–159.
9. Lichtenthaler SF, Wang R, Grimm H, Uljon SN, Masters CL, Beyreuther K. Mech-
anism of the cleavage specificity of Alzheimer’s disease gamma-secretase identi-
fied by phenylalanine-scanning mutagenesis of the transmembrane domain of the
amyloid precursor protein. Proc Natl Acad Sci USA. 1999; 96:3053–3058.
10. Sherrington R, Rogaev EI, Liang Y, et al. Cloning of a gene bearing missense
mutations in early-onset familial Alzheimer’s disease. Nature. 1995; 375:754–
760.
11. Levy-Lahad E, Wijsman EM, Nemens E, et al. A familial Alzheimer’s disease locus
on chromosome 1. Science. 1995; 269:970–973.
12. Kwok JBJ, Taddei K, Hallupp M, et al. Martins RN. Two novel (M233T and
R278T) presenilin-1 mutations in early-onset Alzheimer’s disease pedigrees and
preliminary evidence for association of presenilin-1 mutations with a novel phe-
notype. NeuroReport. 1997; 8:1537–1542.
184 THE AGEING BRAIN

13. Hutton M. Presenilin mutation database. Available from: URL: http://

www.alzforum.org.
14. Murayama O, Tomita T, Nihonmatsu N, et al. Enhancement of amyloid beta 42
secretion by 28 different presenilin 1 mutations of familial Alzheimer’s disease.
Neurosci Lett. 1999; 265: 61–63.
15. Citron M, Westaway D, Xia W, et al. Mutant presenilins of Alzheimer’s disease
increase production of 42-residue amyloid beta-protein in both transfected cells
and transgenic mice. Nat Med. 1997; 3:67–72.
16. Scheuner D, Eckman C, Jensen M, et al. Secreted amyloid beta-protein similar
to that in the senile plaques of Alzheimer’s disease is increased in vivo by the
presenilin 1 and 2 and APP mutations linked to familial Alzheimer’s disease. Nat
Med. 1996; 2:864–870.
17. Sisodia SS, Annaert W, Kim S-H, De Strooper B. Gamma-secretase:never more
enigmatic. Trends Neurosci. 2001; 24 (Suppl): 2–6.
18. Wolfe MS, Xia W, Ostaszewski BL, Diehl TS, Kimberly WT, Selkoe DJ. Two
transmembrane aspartates in presenilin-1 required for presenilin endoproteolysis
and gamma-secretase activity. Nature. 1999; 398:513–517.
19. Li YM, Xu M, Lai MT, et al. Photoactivated gamma-secretase inhibitors directed
to the active site covalently label presenilin 1. Nature. 2000; 405:689–694.
20. Esler WP, Kimberly WT, Ostaszewski BL et al. Transition-state analogue inhibi-
tors of gamma-secretase bind directly to presenilin-1. Nat Cell Biol. 2000; 2:
428–434.
21. Lemere CA, Lopera F, Kosik KS, et al. The E280A presenilin 1 Alzheimer muta-
tion produces increased A beta 42 deposition and severe cerebellar pathology.
Nat Med. 1996; 2:1146–1150.
22. Smith MJ, Gardner RJ, Knight MA, et al. Early-onset Alzheimer’s disease caused
by a novel mutation at codon 219 of the presenilin-1 gene. NeuroReport. 1999;
10: 503–507.
23. Singleton AB, Hall R, Ballard CG, et al. Pathology of early-onset Alzheimer’s
disease cases bearing the Thr113-114ins presenilin-1 mutation. Brain. 2000; 123:
2467–2474.
24. Crook R, Verkkoniemi A, Perez-Tur J, et al. A variant of Alzheimer’s disease with
spastic paraparesis and unusual plaques due to deletion of exon 9 of presenilin 1.
Nat Med. 1998; 4:452–455.
25. Smith MJ, Kwok JBJ, McLean CA, et al. Variable phenotype of Alzheimer’s dis-
ease with spastic paraparesis. Ann Neurol. 2001; 49:125–129.
26. Steiner H, Revesz T, Neumann M, et al. A pathogenic presenilin-1 deletion causes
abberrant Abeta 42 production in the absence of congophilic amyloid plaques.
J Biol Chem. 2001; 276:7233–7239.
27. Houlden H, Baker M, McGowan E, et al. Variant Alzheimer’s disease with
spastic paraparesis and cotton wool plaques is caused by PS-1 mutations that
lead to exceptionally high amyloid-beta concentrations. Ann Neurol. 2000; 48:
806–808.
28. Sato S, Kamino K, Miki T, et al. Splicing mutation of presenilin-1 gene for early-
onset familial Alzheimer’s disease. Hum Mutat. 1998; (Suppl 1): 91–94.
29. Prihar G, Verkkoniem A, Perez-Tur J, et al. Alzheimer disease PS-1 exon 9 dele-
tion defined. Nat Med. 1999; 5:1090.
30. Hiltunen M, Helisalmi S, Mannermaa A, et al. Identification of a novel 4.6-kb
genomic deletion in presenilin-1 gene which results in exclusion of exon 9 in a
Finnish early onset Alzheimer’s disease family: an Alu core sequence-stimulated
recombination. Eur J Hum Genet. 2000; 8:259–266.
31. Rohan de Silva HA, Patel AJ. Presenilins and early-onset familial Alzheimer’s
disease. Neuroreport. 1997; 8:i–xii.
ALZHEIMER’S DISEASE AND FRONTOTEMPORAL DEMENTIA 185

32. Axelman K, Basun H, Lannfelt L. Wide range of disease onset in a family with
Alzheimer disease and a His163Tyr mutation in the presenilin-1 gene. Arch Neu-
rol. 1998; 55:698–702.
33. Tunstall N, Owen MJ, Williams J, et al. Familial influence on variation in age of
onset and behavioural phenotype in Alzheimer’s disease. Brit J Psychiat. 2000;
176:156–159.
34. Silverman JM, Smith CJ, Marin DB, et al. Identifying families with likely genetic
protective factors against Alzheimer disease. Am J Hum Genet. 1999; 64:832–
838.
35. Chartier-Harlin MC, Parfitt M, Legrain S, et al. Apolipoprotein E, epsilon 4
allele as a major risk factor for sporadic early and late-onset forms of Alzheimer’s
disease: analysis of the 19q13.2 chromosomal region. Hum Mol Genet. 1994; 3:
569–574.
36. Corder EH, Saunders AM, Risch NJ, et al. Protective effect of apolipoprotein E
type 2 allele for late onset Alzheimer disease. Nat Genet. 1994; 7:180–184.
37. Laws SM, Taddei K, Fisher C, et al. Evidence that the butylcholinesterase K vari-
ant can protect against late-onset Alzheimer’s disease. Alzheimers Rep. 1999; 2:
219–223.
38. Foster NL, Wilhelmsen K, Sima AA, Jones MZ, D’Amato CJ, Gilman S. Fron-
totemporal dementia and parkinsonism linked to chromosome 17: a consensus
conference. Conference Participants. Ann Neurol. 1997; 41:706–715.
39. Spillantini MG, Goedert M. Tau mutations in familial frontotemporal dementia.
Brain. 2000; 123:857–859.
40. Hutton M, Lendon CL, Rizzu P, et al. Association of missense and 5’-splice-
site mutations in tau with the inherited dementia FTDP-17. Nature. 1998; 393:
702–705.
41. Poorkaj P, Bird TD, Wijsman E, et al. Tau is a candidate gene for chromosome
17 frontotemporal dementia. Ann Neurol. 1998; 43:815–825.
42. Spillantini MG, Murrell JR, Goedert M, Farlow MR, Klug A, Ghetti B. Mutation
in the tau gene in familial multiple system tauopathy with presenile dementia.
Proc Natl Acad Sci USA. 1998; 95:7737–7741.
43. Stanford PM, Halliday GM, Brooks WS, et al. Progressive supranuclear palsy
pathology caused by a novel silent mutation in exon 10 of the tau gene: expan-
sion of the disease phenotype caused by tau gene mutations. Brain. 2000; 123:
880–893.
44. Gotz J, Chen F, van Dorpe J, Nitsch RM. Formation of neurofibrillary tangles
in P301l tau transgenic mice induced by Abeta 42 fibrils. Science. 2001; 293:
1491–1495.
45. Marx J. Neuroscience. New leads on the ‘how’ of Alzheimer’s. Science. 2001;
293: 2192–2194.
46. Wolozin B, Iwasaki K, Vito P, et al. Participation of presenilin 2 in apoptosis:
enhanced basal activity conferred by an Alzheimer mutation. Science. 1996; 274:
1710–1713.
47. Guo Q, Sopher BL, Furukawa K, et al. Alzheimer’s presenilin mutation sensi-
tizes neural cells to apoptosis induced by trophic factor withdrawal and amyloid
beta-peptide: involvement of calcium and oxyradicals. J Neurosci. 1997; 17:
4212–4222.
48. Furukawa K, D’Souza I, Crudder CH, et al. Pro-apoptotic effects of tau mutations
in chromosome 17 frontotemporal dementia and parkinsonism. NeuroReport.
2000; 11:57–60.
Chapter 11

OXIDATIVE AND FREE

RADICAL MECHANISMS
IN BRAIN AGEING
Judy B. de Haan*, Rocco C. Iannello,
Peter J. Crack, Paul Hertzog, and Ismail Kola

Introduction

In this chapter, we discuss the relationship between increased oxidative stress

and cellular ageing, with particular emphasis on the physiological and patho-
logical changes associated with ageing of the brain. In this regard, focus will
be on the role of the antioxidant enzymes during cellular ageing, and the con-
sequences of an altered antioxidant balance. We will highlight the role that
antioxidant genes play in the regulation of senescent-like changes both in in
vitro and in vivo models, and how perturbations of the antioxidant pathways
may lead to clinical outcomes associated with ageing, e.g. neurodegenerative
diseases such as Parkinson’s and Alzheimer’s disease. Furthermore, this chap-
ter will illustrate how an altered antioxidant ratio as a direct consequence of
the over-expression of the antioxidant gene, Sod1, leads to ageing changes
associated with the Down syndrome (DS) phenotype. In this manner it is
hoped to emphasize the importance of the antioxidant genes in the regulation
of redox status during cellular ageing, and how perturbations of redox bal-
ance may have pathological consequences associated with ageing.

Molecular oxygen: a paradox for aerobic organisms

The existence of reactive oxygen species (ROS) within cells is an unavoidable
consequence of both oxidative metabolism and exposure to environmental
*To whom correspondence should be addressed.
188 THE AGEING BRAIN

stresses such as radiation, air pollutants and herbicides. During oxidative

metabolism, oxygen is reduced to water via reactive intermediates that include
the superoxide radical (.O2-), hydrogen peroxide (H2O2), and hydroxyl radi-
cal (.OH). Irrespective of the mode of radical generation, ROS cause cellular
damage through interactions with macromolecules, resulting in mutations in
DNA (both mitochondrial1 and nuclear DNA), destruction of protein struc-
ture and function, and peroxidation of membrane lipids.2
Together with non-enzymatic antioxidants (e.g. ascorbate, glutathione, α-
tocopherol), aerobic organisms have evolved highly efficient enzymatic anti-
oxidant defences to overcome the problems of oxidative stress. These include
the superoxide dismutases (Sod), glutathione peroxidase (Gpx) and catalase
enzymes. Superoxide dismutases function in the first step of the antioxidant
pathway (Figure 1) where .O2- is converted to H2O2, while Gpx and Cat are
independently involved in the neutralisation of H2O2 to water in a second
step. This is a finely tuned process and a balance exists within cells between
the first (Sod) and second steps (Gpx and/or Cat) of the antioxidant pathway.
Indeed it has been postulated that perturbations of this balance could affect
cell function, since a shift in favour of H2O2 (the intermediate product) could
result in Fenton-type reactions with transition metals, resulting in even more
noxious .OH radicals.3 Once formed, these species quickly interact with
molecules in their immediate vicinity, particularly lipids, causing large-scale
macromolecular damage.4

Figure 1. Two-step Antioxidant Pathway. Superoxide radicals generated during oxi-

dative metabolism, are neutralised to water via a two-step process involving
superoxide dismutase (Sod) in a first step, and both or either glutathione
peroxidase (Gpx) and catalase (Cat) in a second step. Fenton-type reactions
occur when an imbalance in this pathway favors the build-up of hydrogen
peroxide (H2O2), resulting in peroxidation of molecules such as lipids.
Adapted from Groner et al.4
OXIDATIVE AND FREE RADICAL MECHANISMS IN BRAIN AGEING 189

Compared with other organs, the brain is most vulnerable to ROS-induced

damage. This is in part due to: (1) the large generation of ROS during oxida-
tive phosphorylation as a consequence of the high rate of oxygen consump-
tion by the brain; (2) high levels of iron in some brain regions which catalyse
the generation of ROS via Fenton-type reactions; and (3) the brain is lipid
rich, particularly in polyunsaturated fatty acids that are known targets of
lipid peroxidation. Thus the levels of cellular antioxidant enzymes become
important during the ageing process, since antioxidants are able to limit
ROS reactions that would otherwise damage macromolecules in an unabated
fashion. Indeed, it is conceivable that continued damage to macromolecules
could translate at a higher level into cellular and/or organ dysfunction and
ultimately cellular ageing and/or death.

Antioxidants and Lipid Peroxidation during Brain Ageing

To gain a better understanding of the role played by antioxidants in limiting

damage during brain ageing, a number of studies have focussed on the activity
of these antioxidants during cellular ageing. However, the literature is con-
troversial with respect to the activities of the major antioxidants, often due
to limitations in the methodologies. Often no distinction is made between the
different isoforms of the superoxide dismutases, such that Sod1 (the cytosolic
and most abundant isoform) and Sod2 (the mitochondrial isoform) are assayed
as one. This has resulted in reports of either an increase,5 a decrease6 or no
change7 in the Sod activity in ageing murine or rat brains. Even when Sod1
and Sod2 are assayed independently, Sod1 has been reported to increase,8
remain constant,9 or decrease10,11 with ageing in mouse or rat brains. Like-
wise, Gpx1 (the most abundant cytosolic and mitochondrial isoform) activity
has been reported to increase5,11 or remain relatively unchanged6,7 in ageing
murine or rat brains. As a marker of oxidative damage in the ageing brain,
the levels of lipid peroxidation are often assayed. Here again, the results have
been controversial, with reports of either an increase11–13 or a decrease14
occurring. Indeed, very few studies simultaneously investigate the activities
of Sod1, Gpx1, catalase and the lipid peroxidation status, which becomes
important if discrepancies such as differences in sex, species, strain and age
are to be eliminated.
We have examined the levels of the three major antioxidant enzymes
(Sod1, Gpx1 and Cat) and the levels of lipid peroxidation in a range of age-
ing murine brains. We show that Sod1 mRNA levels and activity are signifi-
cantly increased in murine brains during the ageing process. Similarly, Sod1
mRNA levels and activity are increased in other murine organs with age.
These include the ageing liver, lung, kidney, heart, ovary, and bone.15–17 An
analysis of the Gpx1 profile showed that most organs adapt to the increased
Sod1 activity by upregulating Gpx1 and/or Cat, at both the mRNA and activ-
ity level with advanced age. However and most importantly, the brain failed
190 THE AGEING BRAIN

to show an increase in either Gpx1 or Cat mRNA or activity with increasing

age. This implies that the ageing brain has an altered Sod1 to Gpx1 and Cat
ratio. As already suggested, an altered ratio can lead to increased H2O2 and
.OH, which in turn can damage macromolecules such as lipids. Indeed the

ageing brain showed significantly increased levels of lipid peroxidation, while

those organs that adapted to the increased Sod1 levels by upregulating Gpx1
and/or catalase showed reduced peroxidative damage.15,17
Our data therefore suggest that an altered antioxidant balance may result
in peroxidative damage to biologically important molecules in the ageing
brain. Regulation of the antioxidant genes during ageing becomes important,
since we are able to show that upregulation of the Sod1 gene in murine brains
is transcriptionally regulated. Why the Gpx1 gene is not upregulated during
murine brain ageing, but is upregulated in other ageing murine organs, is not
yet understood. It may be the need for increased levels of the intermediate
H2O2, to perform other regulatory functions in the brain during the lifetime
of the individual, that out-ways the damaging peroxidative changes seen in
ageing brains. In a recent study, providing the first profile of gene regula-
tion at the molecular level in ageing murine brains, Lee et al. 18 analysed
the expression patterns of 6,347 genes using oligonucleotide array analysis.
They were able to show that ageing of the murine neocortex and cerebellum
resulted in a gene-expression profile indicative of increased oxidative stress,
an altered inflammatory response, and reduced neurotrophic support. These
data provide good evidence at a global level that oxidative stress and inflam-
matory processes, the latter known to involve ROS, are increased in the age-
ing brain and that these processes are regulated at the gene level during brain
ageing.

Lessons from Gene Targeting Experiments

The above studies, which imply a role for an altered antioxidant balance,
increased oxidative stress and peroxidative damage during cellular ageing,
although informative, are still limited since they are of a correlative nature.
To overcome these limitations, and to directly prove a role for ROS in cellular
ageing, researchers have used molecular techniques to address these issues.

In vitro models of cellular senescence

Various studies have shown that over-expression of the human Sod1 gene
in human Hela cells, mouse-L cells and rat PC12 cells, leads to increased
lipid peroxidation as well as structural and functional alterations of lipid
membranes.4,19 We have extended these studies by investigating the effects
of Sod1 overexpression in murine NIH 3T3 cells with respect to cellular
senescence.20
Two types of cell clones were isolated; those overexpressing both Sod1
and Gpx1 which were termed “adapted cell clones”, and those that failed
OXIDATIVE AND FREE RADICAL MECHANISMS IN BRAIN AGEING 191

to upregulate Gpx1 in response to increased Sod1 levels, which were termed

“non-adapted cell clones”. The adapted cells were morphologically and bio-
chemically indistinguishable from the parental lines from which they were
derived. An important observation was that the non-adapted cells looked and
behaved like senescent cells in culture. They exhibited a greater cell-surface
area and had a larger nuclear and cellular volume. These cells also grew more
slowly, as assessed by both growth rate and 3H-dThymidine incorporation.
Furthermore, expression levels of Cip1, which is a well-characterized marker
of cellular senescence, were elevated in all non-adapted cells investigated.20
Since H2O2 is the intermediate between the two steps of the antioxidant
pathway, we measured both intra-and extracellular H2O2 levels of paren-
tal, adapted and non-adapted cells. Hydrogen peroxide levels were not sig-
nificantly different in adapted and parental cells. However, both intra- and
extracellular H2O2 levels were significantly increased in non-adapted cells
compared with both adapted and parental cells. Thus by altering the Sod1
to Gpx1 ratio in favour of H2O2 formation, we were able to demonstrate
senescence-like changes, and we postulated that it was the overproduction of
H2O2, either directly or via Fenton-like reactions to produce .OH radicals,
that was mediating the senescence-like effects. Indeed, we were able to mimic
these changes in both NIH 3T3 cells and primary cultures after the addition
of H2O2.20,21
We addressed the issue of an altered redox balance and senescence-like
changes in two further culture systems where ROS production is elevated as
a consequence of a change in antioxidant gene expression. First we analysed
cultured fibroblasts derived from mice that were genetically manipulated to
have a null mutation for the Gpx1 gene. Indeed, fibroblasts lacking any func-
tional Gpx1 (Gpx1-/- cells) exhibited features of senescence when compared
with control cells, and were more susceptible to H2O2 -induced apoptosis
than controls.
Furthermore, Gpx1-/- neurons demonstrated decreased viability after
exposure to H2O2 compared with controls.22 Second, we showed that cell
lines derived from Down syndrome aborted conceptuses (where the Sod1 to
Gpx1 ratio is increased as a consequence of three copies of the Sod1 gene)
exhibit senescence-like characteristics, namely they grew more slowly, incor-
porated less 3H-dThymidine and expressed higher levels of the senescence
marker, Cip1. Indeed, premature ageing is one of the phenotypes associated
with individuals with DS (see Down syndrome section below). From these
studies, it is evident that an altered antioxidant balance, either as a conse-
quence of the overexpression of Sod1 or a reduction in the activity of Gpx1,
leads to senescence-like changes in cultured cells.

In vivo models of cellular senescence

The most striking data that an altered antioxidant ratio is involved in the
genesis of ageing comes from Yarom et al.,23 who show that mice transgenic
for Sod1, develop morphological and biochemical changes at neuromus-
192 THE AGEING BRAIN

cular junctions of tongue muscles (namely, withdrawal and destruction

of some terminal axons and the development of multiple small terminals),
which are similar to those seen in tongue muscles of ageing mice and rats.24
Furthermore, these changes are similar to those seen in tongue muscles of
individuals with DS.25,26 A subsequent study has also demonstrated premature
ageing changes in the incidence, length and number of nerve branch-points in
Sod1 transgenic hind-limb motor-neuron terminals. Again these changes are
analogous to those seen in ageing mice and rat muscles of the hindlimb.27 The
results may also explain how over-expression of Sod1 affects motor neurons in
individuals with DS, resulting in the impairment of central motoric coordination
and generalized hypotonia of joints. These authors also report ageing changes
in thymocyte populations isolated from Sod1-transgenic animals,28 and the
accumulation of the age-related pigment, lipofuscin, in the myofibers of Sod1-
transgenic animals.27 Importantly, Ceballos et al.29 demonstrate increased lipid
peroxidation in the brains of Sod1-transgenic mice.
Taken together, the above data strongly suggest that altered redox status,
as a consequence of Sod1 over-expression, leads to accelerated ageing changes
in these models.

The role of Free Radical Mechanisms in Diseases of the Ageing Brain

Parkinson’s disease
Parkinson’s disease (PD) is characterized by the progressive loss of pigmented
dopamine-containing neurons in the substantia nigra pars compacta of the brain.
The mechanisms involved in the specific targeting of these cells have received
much attention over the past number of years, since an understanding of these
process(es) may facilitate drug design to either reduce or limit such damage.
One theory is that individuals with PD have a defective antioxidant system
that is incapable of removing harmful ROS generated during the oxidation
of dopamine. Monoamineoxidase catalyses the oxidation of dopamine via the
reactive intermediate, 6-hydroxy-dopamine.30 It is during this process that .O2-
and H2O2 are generated.
The Sod1 activity has been reported to increase in the dopamine-containing
neurons of Parkinsonian brains, possibly as a consequence of the increased
.O - flux.31 Interestingly, it is within these same cell-types that Ceballos et al.32
2
demonstrate increased Sod1 activity in aged brains (with a mean age of 83 yrs).
Cellular damage would be limited if second-step antioxidants were increased
concomitantly. However, Gpx1 activity has been reported to remain unchanged
or is reduced in the substantia nigra of individuals with PD.33 Furthermore, no
difference in either catalase or glutathione reductase is seen in the substantia
nigra or basal nucleus of Parkinsonian brains compared with normal brains.34
Interestingly, Sian et al.35 have shown that levels of reduced glutathione are
decreased by approximately 40–50% in the substantia nigra, thus limiting
the efficient removal of hydrogen peroxide by Gpx in these Parkinsonian
OXIDATIVE AND FREE RADICAL MECHANISMS IN BRAIN AGEING 193

brains (reduced glutathione is required as a cofactor by Gpx). Furthermore,

the potential for the formation of harmful .OH radicals also exists, since
this region exhibits an increase in iron content.36 Thus from in vivo studies
it would appear that there is a significant shift in the balance of antioxidant
enzymes in favour of increased H2O2 and .OH production in these brains, which
may account for the increased cell damage seen in this region of Parkinsonian
brains. Indeed lipid peroxidation is elevated in nigral tissue of post mortem
brains from Parkinsonian individuals compared with control brains.37,38
Further strong evidence for an altered antioxidant balance contributing to
PD pathology comes from recent data of Klivenyi et al.39 who demonstrate
that Gpx1 knockout mice are more susceptible to 1-methyl-4-phenyl-
1,2,5,6-tetrahydropyridine (MPTP)–induced PD. Administration of MPTP,
a mitochondrial toxin and known inducer of oxidative stress, resulted in
significantly greater depletions of dopamine in Gpx1 deficient mice compared
with control mice. These data strongly suggest a neuroprotective role for
Gpx1 in the prevention or reduction of PD-like symptoms after challenge with
mitochondrial toxins. In support of this notion, Bensadoun et al.40 have shown
that mice transgenic for Gpx1 are less susceptible to the toxic effects of 6-
hydroxydopamine, which is known to produce H2O2 and superoxide radicals.
Furthermore, Klivenyi et al.39 suggest that Gpx1 protects against neurotoxicity
by detoxifying harmful peroxynitrite radicals (the latter are formed through the
reaction of nitric oxide and superoxide radicals), since inhibitors of neuronal
nitric oxide synthase, enzymes that generate NO, block MPTP neurotoxicity.
However, it should be emphasised that any genetic defect(s) in free radical
scavenging enzymes appear to be compensated for under physiological
conditions, since we22 and others39,41 show no pathology of Gpx1 knockout
mice under physiological conditions. These defects translate into PD pathology
only when Gpx1 knockout mice are exposed to certain environmental factors
or toxins. These toxins may even be produced endogenously. Naoi and
Maruyama42 demonstrate degeneration of dopaminergic neurons by NM(R)Sal,
an endogenous MPTP-like neurotoxin. This mechanism may also hold true for
individuals susceptible to PD, i.e. an altered antioxidant balance, possibly due
to a reduction in second-step antioxidants such as Gpx1, and exposure to
an environmental/endogenous toxin. A possible therapeutic target for PD is
therefore the upregulation of the glutathione/glutathione peroxidase system,
and in this regard, glutathione precusors such as N-acetyl-cysteine (NAC) have
been considered.43

Stroke
Stroke is the leading cause of long-term disability in adults and ranks as the
third leading cause of death after heart disease and cancer. Approximately
80% of all strokes are ischaemic, that is, due to a reduction of blood flow to
certain brain regions caused by blockage of a vessel. This results in oxygen
deprivation to those regions normally supplied by the occluded blood vessel.
Blood flow back into the occluded region (reperfusion) is accompanied by
194 THE AGEING BRAIN

the production of ROS at an enhanced rate. The increased ROS production is

thought to trigger certain molecular pathways leading to necrosis, apoptosis
and neuroinflammation, resulting in subsequent neuronal loss and serious
cognitive and/or motor disturbances.44 Thus the role played by antioxidants
in the removal of these harmful species becomes extremely important in limit-
ing the neuronal damage post-ischemia.
Use has been made of transgenic and knockout mice to address the issue
of the role of antioxidant genes in the pathogenesis of stroke. For exam-
ple, following mid cerebral artery occlusion and subsequent reperfusion,
the infarct volumes of Sod1 knockout mice are significantly increased, 45,46
implying a protective role for Sod1 against ROS such as .O2-. Conversely, the
infarct volume of Sod1 transgenic mice are reduced compared with control
mice.47 This is somewhat surprising since overexpression of Sod1 should
lead to increased levels of H2O2 and in the absence of an adaptive rise by
second-step antioxidants, increased H2O2 levels are known to be cytotoxic.
Thus the mechanism by which Sod1 over-expression confers protection in
these mice is unclear. It may well be that the Gpx1 gene is upregulated in
stroke-related situations in an adaptive response to the increased levels of
H2O2. In further trying to tease out the role of the various antioxidants in
neuroprotection, Weisbrot-Lefkowitz et al. 48 were able to show that mice
transgenic for Gpx1 show a greater level of protection against ischaemia-
reperfusion damage than Sod1 transgenic mice. These results imply that
H2O2 and/or hydroxyl radicals (formed from H2O2 in the Fenton reaction)
are more neurotoxic than superoxide radicals and therefore second-step
antioxidants such as Gpx1 play a far greater neuroprotective role than the
superoxide dismutases. This becomes particularly important when designing
strategies for drug therapy. In agreement with this notion, recent studies in
our laboratory have shown that Gpx1 knockout mice are more susceptible
to ischaemia-reperfusion injury than controls. We show that the severity of
the infarct volume is significantly increased in Gpx1-/- mice compared with
controls. This increase also correlated with an increase in caspase-3 activa-
tion and an elevation in the level of apoptosis in neural cells. 49 Our results
suggest that Gpx1 plays an important role in the protection of neural cells
against the elevated oxidative stress that accompanies ischemia/reperfusion
injury.

Amyotrophic lateral sclerosis (AML)

Amyotrophic lateral sclerosis (AML) is a progressive disorder of motor neu-
rons found in the cortical regions of the brain, brain stem and spinal cord.
Muscular wasting, weakness and fasciculations, spasticity and hyperreflexia
characterise the disease. From the time of onset, patients with ALS survive a
mean period of 3–4 years. Ninety percent of ALS cases occur as a sporadic
event, while the remaining 10% are inherited as an autosomal dominant
trait, with high penetrance after the sixth decade. Whether the disease occurs
sporadically or is inherited, the clinical features in most instances are similar.
OXIDATIVE AND FREE RADICAL MECHANISMS IN BRAIN AGEING 195

The Sod1 gene has been identified as one of the key players associated with
familial forms of amyotrophic lateral sclerosis (FALS).50
The study of Rosen et al.50 demonstrated that single base mutations in the
coding region of the Sod1 gene are associated with FALS (different mutations
are detected in different families). These mutations resulted in amino-acid
substitutions in regions of the Sod1 enzyme that are highly conserved amongst
organisms, suggesting that these sites are important for Sod1 function.
Transgenic mice expressing mutated forms of Sod1 have provided some of
the most informative data regarding the mechanisms involved in human ALS.
Notably, Gurney et al.51 were able to demonstrate ALS-like symptoms in mice
that over-express a human Sod1 mutation, namely progressive paralysis and
motor neuron loss in the spinal cord and brain stem, thus providing proof that
altered Sod1 function can lead to neurodegenerative changes.52 Furthermore, it
has been shown that mutations in Sod1 result in a dominant gain-of-function
that is peroxidase-like, resulting in the increased formation of .OH radicals.53,54
Recently, Cha et al.55 have shown that neuronal nitric oxide synthetase
(nNOS) expression is enhanced in mutant Sod1 transgenic mice, particularly
within astrocytes, implying a role for NO in the genesis of neurotoxic injury.
NO together with superoxide radicals are known to generate highly noxious
peroxynitrate radicals. Furthermore, mutant Sod1 has been shown to facilitate
peroxynitrite-mediated nitration of proteins in mutant Sod1 transgenic mice.56
Pathogenesis in the transgenic mouse model of FALS has recently been
proposed to occur via a two-step sequential process, in which damage is mediated
by free radicals which accumulate to a threshold, triggering catastrophic motor
neuron loss through glutamate-mediated excitotoxic mechanisms.57 Evidence
in support of this hypothesis comes from therapeutic studies with antioxidants
and inhibitors of glutamatergic neurotransmission. Feeding of mutant Sod1
transgenic mice with vitamin E and selenium (selenium is an essential cofactor of
Gpx1) delayed onset of the disease, and strength and mobility were transiently
improved compared with non-supplemented Sod1 mutant mice. Administering
riluzole and gabapentin, two drugs that reduce presynaptic glutamate release or
biosynthsis, improved survival of the mutant Sod1 mice.58 It is probably correct
to assume that both sporadic and familial forms of ALS trigger a common ROS-
mediated pathway of motor neuron death that is .OH and/or peroxynitrate-
mediated.

Alzheimer’s disease
Alzheimer’s disease (AD) affects 7% of the population over 65 years of age
and is characterized by slow progressive intellectual decline and personality
deterioration. Autopsy studies show intra-neuronal fibrillary tangles and neu-
ritic plaques. The latter are spherical extracellular cores of β-amyloid protein
surrounded by degenerating nerve-cell processes. Both plaques and tangles
occur throughout the cerebral cortex of the brain and consist of bundles of
uniform proteins that appear as paired helical filaments on electron micro-
scopic examination.59
196 THE AGEING BRAIN

There is now growing evidence that amyloid β-peptide (Aβ) and oxidative
stress are implicated in the pathogenesis of AD.60,61 It has been shown that Aß
is over-produced in the brains of patients with AD and that Aβ peptides can
be toxic to neuronal cells through a mechanism that involves H2O2.62 Indeed,
oxidative stress actually exacerbates Aβ aggregation,63 while the deposition
of Aβ in turn, increases intraneuronal generation of ROS.64 In this manner,
a cyclical response is established with continued deposition of Aβ. In vitro
studies have also shown marked oxidative injury, including lipid peroxidation,
protein carbonyl formation, mitochondrial DNA damage and the induction of
stress related alterations that include ubiquination of cytoskeletal proteins.65
Furthermore, the amino acid hydroxyproline that is not normally a constituent
of cytoplasmic protein in the brain, was identified as an integral part of paired
helical filament proteins in AD brains. This led Zemlan et al.59 to propose that
these modified amino acids arise due to non-enzymatic hydroxylation of proline
residues, presumably arising from .OH radicals. If the hypothesis that oxidative
stress contributes to the pathology of AD is correct, then H2O2 and/or .OH
could be elevated in Alzheimer’s disease as a consequence of an alteration in
antioxidant balance.
Indeed, Sod1 levels are elevated in AD brains. In particular, post-mortem
analysis of AD brains showed that large pyramidal neurons of the hippoc-
ampus contained higher amounts of Sod1 mRNA and protein than control
brains.66 It is these cells that are particularly vulnerable to degenerative proc-
esses in AD. In addition, Sod1 activity is increased by 30% in fibroblasts of
familial AD patients compared with normal controls.59 Also lending support
to this notion are the data from amyloid precursor protein (APP)-transgenic
mice that show elevated Sod1 levels in brain regions, especially around Aβ
deposits.65 These authors also demonstrate an increase in heme-oxygenase-
1, a marker of oxidative stress, around most amyloid deposits in their APP-
transgenic mice. Further evidence that oxidative damage may contribute to
cellular damage in AD brains comes from an analysis of lipid peroxidation
in these brains. Indeed, the two regions most susceptible to neurodegenera-
tion in AD, the temporal and parietal cortex, showed elevated levels of lipid
peroxidation, whilst the least affected areas, namely the occipital cortex and
cerebellum, showed no elevation when compared with control brains.67

Overexpression of the Antioxidant Gene Sod1, Leads to Ageing Changes

Associated with the Down Syndrome Phenotype

Accelerated ageing changes have been observed in individuals with Down syn-
drome. In particular, the rapid or early onset of ageing is evident visually as
premature greying or loss of hair. Detailed biochemical analysis has revealed
that individuals with DS show a decline in immune responsiveness similar
to that seen in older people. Furthermore, alterations in cyclic nucleotide
metabolism have been noted in lymphocytes from individuals with DS, which
OXIDATIVE AND FREE RADICAL MECHANISMS IN BRAIN AGEING 197

is comparable with what occurs in ageing human lymphocytes. Granulo-

vacuolar degeneration of neurons and the appearance of Alzheimer’s disease
pathology, amyloidosis, hypogonadism and degenerative vascular disease
have also been noted in DS individuals.68,69
A perturbation in the ratio of Sod1 relative to Gpx1 and/or catalase is of
particular relevance in DS. Importantly, the Sod1 gene is located on human
Ch21, and in 95% of cases the entire chromosome is triplicated in DS
individuals. Indeed, elevated Sod1 activity has been observed in tissues such
as red blood cells, platelets, fibroblasts and lymphocytes of DS individuals.70
Furthermore, we show that the Sod1 to Gpx1 ratio is increased in five fetal DS
organs.16 These include the fetal brain, liver, thymus, lung and heart. Our results
are in agreement with Brooksbank and Balazs71 who demonstrate increased
Sod1 activity, unaltered Gpx activity, and increased lipid peroxidation in DS
fetal brains. However an altered antioxidant balance may not occur in all DS
organs and tissues, since a compensatory increase in Gpx1 activity occurs in
erythrocytes and lymphoid cells of DS individuals.72 Interestingly, catalase
activity has been shown to be unaffected in DS erythrocytes.73 Thus it appears
that the ratio of the major antioxidant enzymes is shifted in favour of H2O2
production in the majority of organs, and may in this manner contribute to the
age-related pathologies associated with DS.74 Evidence in favour of this hypoth-
esis comes from data of Buscioglio and Yankner75 who demonstrate increased
ROS formation in primary cultures of DS fetal neurons. Glycoxidation and in
particular, the accumulation of advanced glycation end products (AGEs), are
also enhanced in DS fetal brains.76 These results suggest that brain oxidative
stress occurs very early in the life of DS individuals, and may contribute to
some of the pathologies associated with DS, such as the premature ageing and
neurodegenerative AD-like pathology.
Of particular interest has been the age-associated neuritic plaque for-
mation analogous to that found in AD.77 Post-mortem examination of DS
brains over the age of 35 years, almost invariably show pathology similar to
that seen in AD. Interestingly, the areas most affected in DS brains parallel
those affected in individuals with AD. Evidence has already been presented
that ROS may contribute to the pathology of AD. Indeed, the gene dosage
increase in Sod1 activity in DS may contribute in a similar fashion to the AD-
like pathology. However, in this instance it likely to be the concordant effect
of increased Sod1 activity and increased Aβ production that is responsible for
the AD-like pathology in DS, since the gene coding for APP is also localized to
human chromosome 21 and is over-expressed in DS.78

Conclusion

This chapter has described the role of the antioxidant genes and their gene-prod-
ucts in the regulation of cellular ageing. It has focussed primarily on their role in
brain ageing since this organ is particularly vulnerable to peroxidative insults.
198 THE AGEING BRAIN

Figure 2. An altered antioxidant balance may have pathological consequences. An

imbalance in the major antioxidants can result in the build-up of noxious
radicals, predisposing and/or contributing to various pathologies, e.g. (i)
neuropathological outcomes such as stroke, Familial Lateral Sclerosis
(FALS), Parkinson’s and Alzheimer’s disease, (ii) Down syndrome as a con-
sequence of the overexpression of SOD1, a chromosome 21 gene, and (iii)
the ageing process per se, which in turn may predispose to various patholo-
gies. Furthermore inflammation often accompanies these pathologies e.g.
AD, PD and Down syndrome, often as a consequence of radical-mediated
induction of transcription factors such as NF-κB, leading to upregulation
of TFN-α, IL-1 and IL-6. This in turn generates more radicals. An increase
in both SOD1 and APP may contribute to AD-like pathology in DS.

It has shown that an altered Sod1 to Gpx1 and catalase ratio exists in ageing
murine brains and that this altered ratio is accompanied by an increase in lipid
peroxidation. It has highlighted the importance of maintaining a redox balance
in cells and that a perturbation in first to second step antioxidant enzymes can
affect cell function, leading to senescence-like changes. Furthermore, evidence
was presented that altered redox states exist in various pathologies associated
with ageing (Figure 2).
Cumulative evidence now strongly suggests the existence of a molecular
basis for ageing, with the regulation of the antioxidant genes playing an impor-
tant role in this regard. Current thinking supports the hypothesis of Sohal and
Allen79 who extended the free radical theory of ageing (which was based on the
production of ROS in an uncontrolled fashion during aerobic metabolism), to
OXIDATIVE AND FREE RADICAL MECHANISMS IN BRAIN AGEING 199

include controlled genetic changes induced by ROS during ageing. It is only by

clearly defining the molecular basis of senescence-like changes during normal
cellular ageing and in pathologies associated with ageing, that one can hope to
design drugs to reduce or ameliorate these detrimental ROS-induced effects.

References

1. Richter C. Oxidative damage to mitochondrial DNA and its relationship to age-

ing. Int J Biochem Cell Biol. 1995; 27:647–653.
2. Halliwell B, Gutteridge JMC. In: Free radicals in biology and medicine, Oxford,
Clarendon Press, 1985.
3. Imlay JA, Chin SM, Linn S. Toxic DNA damage by hydrogen peroxide through
the Fenton reaction in vivo and in vitro. Science. 1988; 240:640–642.
4. Groner Y, Elroy-Stein O, Avraham KB, Yarom R, Schickler M, Knobler H, Rot-
man G. Down syndrome clinical symptoms are manifested in transfected cells and
transgenic mice overexpressing the human Cu/Zn-superoxide dismutase gene. J
Physiology-Paris. 1990; 84:53–77.
5. Tayarani I, Cloez I, Clement M, Bourre JM. Antioxidant enzymes and related
trace elements in ageing brain capillaries and choroid plexus. J Neurochem. 1989;
53:817–824.
6. Benzi G, Pastoris O, Villa RF. Changes induced by ageing and drug treatment on
cerebral enzymatic antioxidant system. Neurochem Res. 1988; 13:467–478.
7. Cand F, Verdetti J. Superoxide dismutase, glutathione peroxidase, catalase, and
lipid peroxidation in the major organs of the ageing rats. Free Radical Bio Med.
1989; 7:59–63.
8. Bracco F, Burlina AP, Malesani R, Rigo A, Battistin L. Free-radical related
enzymes in the ageing brain. In: Bes A, et al. editors. Senile dementias: Early
detection, Libbey, 1986; 293–297.
9. Kurobe N, Suzuki F, Kato K, Sato T. Sensitive immunoassay of rat Cu/Zn super-
oxide dismutase: concentrations in the brain, liver, and kidney are not affected
by ageing. Biomed Res. 1990; 11:187–194.
10. Mariucci G, Ambrosini MV, Colarieti L, Bruschelli G. Differential changes in
Cu, Zn and Mn superoxide dismutase activity in developing rat brain and liver.
Experientia 1990; 46:753–755.
11. Sahoo A, Chainy GBN. Alterations in the activities of cerebral antioxidant enzymes
of rat are related to ageing. Int J Dev Neurosci. 1997; 15:939–948.
12. Mizuno Y, Ohta K. Regional distributions of thiobarbituric acid-reactive prod-
ucts, activities of enzymes regulating the metabolism of oxygen free radicals, and
some of the related enzymes in adult and aged rat brains. J Neurochem. 1986;
46:1344–1352.
13. Sawada M, Carlson JC. Changes in superoxide radical and lipid peroxide forma-
tion in the brain, heart and liver during the lifetime of the rat. Mech Ageing Dev.
1987; 41:125–137.
14. Boehme DH, Kosecki R, Stern F, Marks N. Lipoperoxidation in human and rat
brain tissue: development and regional studies. Brain Res. 1977; 136:11–21.
15. de Haan JB, Newman JD, Kola I. Cu/Zn superoxide dismutase mRNA and
enzyme activity, and susceptibility to lipid peroxidation, increases with ageing in
murine brains. Mol Brain Res. 1992; 13:179–186.
16. de Haan JB, Wolvetang E, Cristiano F, Iannello R, Kelner M, Kola I. Reactive
oxygen species and their contribution to pathology in Down syndrome. Adv
Pharmacol. 1997; 38:379–402.
200 THE AGEING BRAIN

17. Cristiano F, de Haan JB, Iannello I, Kola I. Changes in the levels of enzymes which
modulate the antioxidant balance occur during ageing and correlate with cellular
damage. Mech Ageing Dev. 1995; 80:93–105.
18. Lee C-K, Weindruch R, Prolla TA. Gene-expression profile of the ageing brain in
mice Nat Genet. 2000; 25:294–297.
19. Elroy-Stein O, Bernstein Y, Groner Y. Overproduction of human Cu/Zn-superox-
ide dismutase in transfected cells: extenuation of paraquat-mediated cytotoxicity
and enhancement of lipid peroxidation. EMBO J. 1986; 5:615–622.
20. de Haan JB, Cristiano F, Iannello R, Kelner M, Kola I. Elevation in the ratio of
Cu/Zn-superoxide dismutase to glutathione peroxidase leads to cellular senes-
cence and this effect is mediated by H2O2. Hum Mol Genet. 1996; 5:283–292.
21. Bladier C, Wolvetang EJ, Hutchinson P, de Haan JB, Kola I. Response of a
primary human fibroblast cell line to H 2O2: Senescence-like growth-arrest or
apoptosis? Cell Growth Differ. 1997; 8:589–598.
22. de Haan JB, Bladier C, Griffiths P, Kelner M, O’Shea RD, Cheung NS, Bronson
RT, Silvestro M.J, Wild S, Zheng SS, Beart PM, Hertzog PJ, Kola I. Mice with a
homozygous null mutation for the most abundant glutathione peroxidase Gpx1,
show increased susceptibility to the oxidative stress-inducing agents paraquat and
hydrogen peroxide. J Biol Chem. 1998; 273:22528–22536.
23. Yarom R, Sapoznikov D, Havivi Y, Avraham KB, Schickler M, Groner Y. Prema-
ture ageing changes in neuromuscular junctions of transgenic mice with an extra
human CuZnSOD gene: a model for tongue pathology in Down’s Syndrome. J
Neurol Sci. 1988; 88:41–53.
24. Fahim MA, Robbins N. Ultrastructural studies of young and old mouse-neu-
romuscular junctions. J Neurocytol. 1982; 11:641–656.
25. Yarom R, Sherman Y, Sagher U, Peled IJ, Wexler MR. Elevated concentrations
of elements and abnormalities of neuromuscular junctions in tongue muscles of
Down’s syndrome. J Neurol Sci. 1987; 79:315–326.
26. Epstein CJ, Avraham KB, Lovett M, Smith S, Elroy-Stein O, Rotman G, Bry C,
Groner Y. Transgenic mice with increased Cu/Zn-superoxide dismutase activity:
Animal model of dosage effects in Down syndrome. Proc Natl Acad Sci USA.
1987; 84:8044–8048.
27. Avraham KB, Sugarman H, Rotshenker S, Groner Y. Down’s syndrome: morpho-
logical remodelling and increased complexity in the neuromuscular junction of
transgenic CuZn-superoxide dismutase mice. J Neurocytol. 1991; 20:208–215.
28. Peled-Kamar M, Lotem J, Okon E, Sachs L, Groner Y. Thymic abnormalities
and enhanced apoptosis of thymocytes and bone marrow cells in transgenic mice
over-expressing Cu/Zn-superoxide dismutase: implications for Down syndrome.
EMBO J. 1995; 14:4985–4993.
29. Ceballos I, Nicole A, Briand P, Grimber G, Delacourte A, Flament S, Blouin JL,
Thevenin M, Kamoun P, Sinet PM. Expression of human Cu-Zn superoxide dis-
mutase gene in transgenic mice: model for gene dosage effect in Down syndrome.
Free Radical Res Com. 1991; 12-13:581–589.
30. Cohen G. The pathobiology of Parkinson’s disease: biochemical aspects of
dopamine neuron senescence. J Neural Transm Supp. 1983; 19:89–103.
31. Saggu H, Cooksey J, Dexter D, Wells FR, Lees A, Jenner P, Marsden CD. A
selective increase in particulate superoxide dismutase activity in parkinsonian
substantia nigra. J Neurochem. 1989; 53:692–697.
32. Ceballos I, Lafron M, Javoy-Agid F, Hirsch E, Sinet PM, Agid Y. Superoxide
dismutase and Parkinson’s disease. Lancet. 1990; 335:1035–1036.
33. Kish SJ, Morito C, Hornykiewicz O. Glutathione peroxidase activity in Parkin-
son’s disease brain. Neurosci Lett. 1985; 58:343–346.
34. Marttila RJ, Lorentz H, Rinne UK. Oxygen toxicity protecting enzymes in Par-
kinson’s disease. J Neurol Sci. 1988; 86:321–331.
OXIDATIVE AND FREE RADICAL MECHANISMS IN BRAIN AGEING 201

35. Sian J, Dexter DT, Lees AJ, Daniel S, Agid Y, Javoy-Agid F, Jenner P, Marsden
CD. Alterations in glutathione levels in Parkinson’s disease and other neurode-
generative disorders affecting basal ganglia. Ann Neurol. 1994; 36:348–355.
36. Hirsch EC, Faucheux BA. Iron metabolism and Parkinson’s disease. Movement
Disord. 1998; 13:39–45.
37. Dexter DT, Carter G, Agid F, Agid Y, Lees AJ, Jenner P, Marsden CD. Lipid
peroxidation as cause of nigral cell death in Parkinson’s disease. Lancet. 1986;
11:639–640.
38. Jenner P, Dexter DT, Sian J, Schapira AH, Marsden CD. Oxidative stress as a
cause of nigral cell death in Parkinson’s disease and incidental Lewy body disease.
Ann Neurol. 1992; 32:S82–87.
39. Klivenyi P, Andreassen OA, Ferrante RJ, Dedeoglu A, Mueller G, Lancelot E,
Bogdanov M, Andersen JK, Jiang D, Beal MF. Mice deficient in cellular glutath-
ione peroxidase show increased vulnerability to malonate, 3-nitropropionic acid,
and 1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine. J Neurosci. 2000; 20:1–7.
40. Bensadoun JC, Mirochnitchenko O, Inouye M, Aebischer P, Zurn AD. Attenu-
ation of 6-OHDA-induced neurotoxicity in glutathione peroxidase transgenic
mice. Eur J Neurosci. 1998; 10:3231–3236.
41. Ho YS, Magnenat JL, Bronson RT, Cao J, Gargano M, Sugawara M, Frank CD.
Mice deficient in cellular glutathione peroxidase develop normally and show no
increased sensitivity to hyperoxia. J Biol Chem. 1997; 272:16644–16651.
42. Naoi M, Maruyama W. Cell death of dopamine neurons in ageing and Parkin-
son’s disease. Mech Ageing Dev. 1999; 111:175–188.
43. Martinez M, Martinez N, Hernandez AI, Ferrandiz ML. Hypothesis: can N-ace-
tylcysteine be beneficial in Parkinson’s disease. Life Sci. 1999; 64:1253–1257.
44. Dirnagl U, Iadecola C, Moskowitz MA. Pathobiology of ischemic stroke: an
integrated view. Trends Neurosci. 1999; 22:391–397.
45. Kondo T, Reaume AG, Huang TT, Carlson E, Murakami K, Chen SF, Hoffman
EK, Scott RW, Epstein CJ, Chan PH. Reduction of CuZn-superoxide dismutase
activity exacerbates neuronal cell injury and edema formation after transient focal
cerebral ischemia. J Neurosci. 1997; 17:4180–4189.
46. Murakami K, Kondo T, Kawase M, Li Y, Sato S, Chen SF, Chan PH. Mitochon-
drial susceptibility to oxidative stress exacerbates cerebral infarction that follows
permanent focal cerebral ischemia in mutant mice with manganese superoxide
dismutase deficiency. J Neurosci. 1998; 18:205–213.
47. Yang G, Chan PH, Chen J, Carlson E, Chen SF, Weinstein P, Epstein CJ, Kamii
H. Human copper-zinc superoxide dismutase transgenic mice are highly resistant
to reperfusion injury after focal cerebral ischemia. Stroke. 1994; 25:165–170.
48. Weisbrot-Lefkowitz M, Reuhl K, Perry B, Chan PH, Inouye M, Mirochnitchenko
O. Overexpression of human glutathione peroxidase protects transgenic mice
against focal cerebral ischemia/reperfusion damage. Brain Res Mol Brain Res.
1998; 53:333–338.
49. Crack PJ, Taylor JM, Flentjar NJ, de Haan J, Hertzog P, Iannello RC, Kola I.
Increased infarct size and exacerbated apoptosis in the glutathione peroxidase-
1 (Gpx-1) knockout mouse brain in response to ischemia/reperfusion injury.
J Neurochem. 2001; 78:1389–1399.
50. Rosen DR, Siddique T, Patterson D, Figlewicz DA, Sapp P, Hentati A, Donaldson
D, Goto J, O’Regan JP, Deng HX, Rahmani Z, Krizus A, McKenna-Yasek D,
Cayabyab A, Gaston SM, Berger R, Tanzi RE, Halperin JJ, Hertzfeldt B, Van den
Bergh R, Hung WY, Bird T, Deng G, Mulder DW, Smyth C, Laing NG, Soriano
E, Pericak-Vance MA, Haines J, Rouleau GA, Gusella JS, Horvitz HR, Brown RH
Jr. Mutations in Cu/Zn superoxide dismutase gene are associated with familial
amyotrophic lateral sclerosis. Nature. 1993; 362:59–62.
202 THE AGEING BRAIN

51. Gurney ME, Pu H, Chiu AY, Dal Canto MC, Polchow CY, Alexander DD,
Caliendo J, Hentati A, Kwon YW, Deng HX. Motor neuron degeneration in mice
that express a human Cu, Zn superoxide dismutase mutation. Science. 1994; 264:
1772–1775.
52. Dal Canto MC, Gurney ME. A low expressor line of transgenic mice carrying a
mutant human Cu,Zn superoxide dismutase (SOD1) gene develops pathological
changes that most closely resemble those in human amyotrophic lateral sclerosis.
Acta Neuropathol. 1997; 93:537–550.
53. Gurney ME, Cutting FB, Zhai P, Andrus PK, Hall ED. Pathogenic mechanisms
in Familial Amyotrophic Lateral Sclerosis due to mutation of Cu, Zn superoxide
dismutase. Pathol Biol. 1996; 44:51–56.
54. Liu R, Althaus JS, Ellerbrock BR, Becker DA, Gurney ME. Enhanced oxygen
radical production in a transgenic mouse model of familial amyotrophic lateral
sclerosis. Ann Neurol. 1998; 44:763–770.
55. Cha CI, Kim JM, Shin DH, Kim YS, Kim J, Gurney ME, Lee KW. Reactive
astrocytes express nitric oxide synthetase in the spinal cord of transgenic mice
expressing a human Cu/Zn SOD mutation. NeuroReport. 1998; 9:1503–1506.
56. Ferrante RJ, Shinobu LA, Schulz JB, Matthews RT, Thomas CE, Kowall NW,
Gurney ME, Beal MF. Increased 3-nitrotyrosine and oxidative damage in mice
with a human copper/zinc superoxide dismutase mutation. Ann Neurol. 1997;
42:326–334.
57. Gurney ME. Transgenic animal models of familial amyotrophic lateral sclerosis.
J Neurol. 1997; 244:S15–20.
58. Gurney ME, Cutting FB, Zhai P. Benefit of vitamin E, riluzole and gabapentin in
a transgenic model of familial amyotrophic lateral sclerosis. Ann Neurol. 1996;
39:147–157.
59. Zemlan FP, Thienhaus OJ, Bosmann HB. Superoxide dismutase activity in Alzhe-
imer’s disease: possible mechanism for paired helical filament formation. Brain
Res. 1989; 476:160–162.
60. Hensley K, Carney JM, Mattson MP, Aksenova M, Harris M, Wu JF, Floyd RA,
Betterfield DA. A model for beta-amyloid aggregation and neurotoxicity based
on free radical generation by the peptide: relevance to Alzheimer disease. Proc
Natl Acad Sci USA. 1994; 91:3270–3274.
61. Iannello RC, Crack PJ, de Haan JB, Kola I. Oxidative stress and neural dysfunc-
tion in Down syndrome. J Neural Transm. 1999; 57:257–267.
62. Behl C, Davis JB, Lesley R, Schubert D. Hydrogen peroxide mediates amyloid
beta protein toxicity. Cell. 1994; 77:817–827.
63. Multhaup G, Ruppert T, Schlicksupp A, Hesse L, Beher D, Masters CL, Beyreuther
K. Reactive oxygen species and Alheimer’s disease. Biochem Pharmacol. 1997;
54:533–559.
64. Yan SD, Yan SF, Chen X, Fu J, Chen M, Kuppusamy P, Smith MA, Perry G,
Godman GC, Nawroth P, et al. Non-enzymatically glycated tau in Alzheimer’s
disease induces neuronal oxidant stress resulting in cytokine gene expression and
release of amyloid beta-peptide. Nat Med. 1995; 1:693–699.
65. Pappolla MA, Chyan YJ, Omar RA, Hsiao K, Perry G, Smith MA, Bozner P. Evi-
dence of oxidative stress and in vivo neurotoxicity of beta-amyloid in a transgenic
mouse model of Alzheimer’s disease: a chronic oxidative paradigm for testing
antioxidant therapies in vivo. Am J Pathol. 1998; 152:871–877.
66. Furuta A, Price DL, Pardo CA, Troncoso JC, Xu ZS, Taniguchi N, Martin LJ.
Localization of superoxide dismutases in Alzheimer’s Disease and Down’s syn-
drome neocortex and hippocampus. Am J Pathol. 1995; 146:357–367.
67. Hajimohammadreza I, Brammer M. Brain membrane fluidity and lipid peroxida-
tion in Alzheimer’s disease. Neurosci Lett. 1990; 112:333–337.
OXIDATIVE AND FREE RADICAL MECHANISMS IN BRAIN AGEING 203

68. Kola I, Cristiano F, de Haan JB, Sumarsono S, Thomas R, Corrick C, Tymms M.

Genes, Embryogenesis and Down syndrome. In: Moeloek F, Affandi B, Trounson
AO, editors. Advances in human reproduction. Casterton: Parthanon Publishing,
1995; 309–320.
69. Kola I, Hertzog PJ. Animal models in the study of the biological function of
genes on human chromosome 21 and their role in the pathophysiology of Down
syndrome. Hum Mol Genet. 1997; 6:1713–1727.
70. Anneren KG, Epstein CJ. Lipid peroxidation and superoxide dismutase-1 and
glutathione peroxidase activities in trisomy 16 fetal mice and human trisomy 21
fibroblasts. Pediatr Res. 1987; 21:88–92.
71. Brooksbank BWL, Balazs R. Superoxide dismutase, glutathione peroxidase and
lipoperoxidation in Down’s Syndrome fetal brain. Dev Brain Res. 1984; 16:
37–44.
72. Sinet PM, Michelson AM, Bazin A, Lejeune J, Jerome H. Increase in glutathione
peroxidase activity in erythrocytes from trisomy 21 subjects. Biochem Biophys
Res Com. 1975; 67:910–915.
73. Percy ME, Dalton AJ, Markovic VD, Crapper McLachlan DR, Hummel JT,
Rusk ACM, Andrews DF. Red cell superoxide dismutase, glutathione peroxidase
and catalase in Down syndrome patients with and without manifestations of
Alzheimer’s disease. Am J Med Genet. 1990; 35:469–467.
74. Bladier C, de Haan JB, Kola I. Antioxidant genes and reactive oxygen species in
Down syndrome. In: Sen CK, Sies H, Baeuerle PA, editors. Antioxidant and redox
regulation of genes. San Diego: Academic Press, 2000; 425–449.
75. Busciglio J, Yankner BA. Apoptosis and increased generation of reactive oxygen
species in Down’s syndrome neurons in vitro. Nature. 1995; 378:776–779.
76. Odetti P, Angelini G, Dapino D, Zaccheo D, Garibaldi S, Dagna-Bricarelli F,
Piombo G, Perry G, Smith M, Traverso N, Tabaton M. Early glyoxidation dam-
age in brains from Down’s syndrome. Biochem Biophys Res Com. 1998; 243:
849–851.
77. Mann DMA, Esiri MM. The pattern of acquisition of plaques and tangles in the
brains of patients under 50 years of age with Down’s Syndrome. J Neurol Sci.
1989; 89:169–179.
78. Neve RL, Finch EA, Dawes LR. Expression of the Alzheimer amyloid precursor
gene transcripts in the human brain. Neuron 1988; 1:669–677.
79. Sohal RS, Allen RG. Oxidative stress as a causal factor in differentiation and
ageing: a unifying hypothesis. Exp Geront. 1990; 25:499–522.
Chapter 12

THE ROLE OF
NUTRITIONAL FACTORS
IN COGNITIVE AGEING
Janet Bryan

Introduction

The association between nutrition and cognitive performance has become

a topic of increasing scientific and public interest. 1 Nutrition may be an
important, modifiable, life-style factor in age-related cognitive decline and
there is a growing interest in the development of nutritional supplements as
therapeutic agents aimed at enhancing or maintaining cognitive function or
delaying cognitive decline.2 The role that food and nutrition has in the course
of age-related cognitive change is yet to be clearly defined. However, early
findings from epidemiological and experimental studies suggest that there
may be a role for dietary components such as: folate and vitamins B-12 and
B-6, antioxidants, omega 3 fatty acids, and herbal supplements like Ginkgo
biloba.
The assumption that food and its components can impact on cognitive
performance and age-related cognitive change is based on the knowledge that
the central nervous system (CNS) depends heavily for efficient functioning
on a constant supply of almost all of the essential nutrients as well as glucose
and oxygen.3 If brain function is optimised, one would expect cognitive func-
tion to be optimised as well. Understanding this link between brain function
and cognitive function is important in the formulation of hypotheses about
diet-cognition links and how we might test them. We need to base research in
the area of diet and cognition on hypotheses about the mechanisms by which
nutrients or other aspects of the diet might impact on the brain, and how this
in turn affects cognitive performance. Nutrients may impact on the brain in
206 THE AGEING BRAIN

a number of ways such as: on neurotransmitter synthesis, on the structure

of neurons, and on the vasculature of the brain. The effects of nutrients on
the brain may be short-term and acute, or more longer-term due to dietary
habits over an extended period of time. Importantly, if we understand the
mechanisms by which nutrients affect the brain, we can more accurately pre-
dict the role that nutrients may have in cognitive ageing and which aspects of
cognition are likely to be affected by nutritional factors.
At a very basic level, we might expect nutritional factors to impact on
fluid, rather than crystallised, abilities among older adults.4 Fluid ability is
thought to reflect innate information processing that is determined by genetic
and physiological factors such as the integrity of the CNS. Fluid ability is
demonstrated in the capacity to process novel information, that is, to apply
mental processes to situations that require no previous knowledge. Crystal-
lised ability refers to the application of learned information and cultural
experience acquired over the lifetime and therefore depends on our store of
knowledge, education and cultural background. Due to its reliance on the
integrity of the CNS, we might expect nutritional factors to have more of an
impact on fluid, rather than crystallised, ability.
Furthermore, there are some aspects of cognition that show marked age-
related decline, such as memory performance and cognitive resources, that
may also be sensitive to nutritional factors. Cognitive resources refer to our
mental capacity to perform cognitive tasks. The three identified cognitive
resources are: speed of information processing (how fast we can think); work-
ing memory capacity (how much information we can simultaneously store
and work on); and attentional capacity (how long we can concentrate for).
Because these resources are so important for the efficient working of the cog-
nitive system,5 any investigation of the links between nutrition and cognitive
performance should include measures of cognitive resources.4 In summary,
because cognition is multidimensional, it is crucial that pertinent tests of
cognitive performance be selected in order to test hypotheses that specify the
impact that nutritional factors might have on the cognitive system.

Folate, and Vitamins B-12 and B-6

Recent research has focussed on the role of B vitamins, especially folate,

B-12 and B-6, in cognitive ageing.6 Although most of the evidence for the link
between B vitamins and cognitive performance is based on studies involving
participants with clinical vitamin deficiencies, the effect of these vitamins
might also be important for a broader population.3, 7 In particular, a mild to
moderate folate deficiency is thought to be relatively common in the general
population,8 while the incidence of clinical and subclinical levels of folate
and B-12 deficiency has been found to be higher in the elderly.8–12 Stabler et
al.13 estimate that low serum B-12 concentrations might be evident in 5–15%
of the elderly population. They suggest multiple causes of B-12 deficiency in
NUTRITION AND COGNITIVE AGEING 207

older adults such as: pernicious anaemia, atrophic gastritis causing B-12 mal-
absorption, or previous gastric or intestinal surgery. Evidence is continuing to
mount that higher intakes and serum concentrations of certain vitamins, par-
ticularly folate and B-12, might be beneficial for cognitive performance.14
Research to date indicates two inter-linked neurochemical mechanisms by
which these B vitamins influence cognitive performance through their role
in methylation in the CNS.15–17 The first mechanism (the hypomethylation
hypothesis) posits that folate, with B-12 or B-6 as catalysing cofactors, may
have a direct and possibly acute effect on the CNS via hypomethylation,
which inhibits the synthesis of methionine and the formation of S-adenosyl-
methionine (SAMe). This in turn inhibits methylation reactions throughout
the CNS involving proteins, membrane phospholipids, DNA, the metabolism
of neurotransmitters such as the monoamines (e.g. dopamine, norepinephrine,
serotonin), and melatonin, all of which are crucial to neurological and psy-
chological status.16–19 The second mechanism (the homocysteine hypothesis)
proposes that there is an indirect and possibly longer-term effect of folate,
B-12 and B-6 on the functioning of the brain via the cerebrovasculature. Stud-
ies have demonstrated that high levels of homocysteine, largely attributable
to low levels of folate, and B-12 or B-6, are associated with increased risk of
vascular disease14,15,20–22 due to toxic effects on vascular tissue, or excessive
production of excitotoxic sulphur amino acids, homocysteic acid and cystein
sulphinic acid.11 Thus, these B vitamins may function to preserve the integrity
of the CNS via their role in the prevention of vascular disease, which is crucial
to cognitive function.15,16,23–25

Cross-sectional studies
Most studies investigating the links between the B vitamins and cognitive per-
formance among older adults have employed correlational designs. Goodwin
et al.26 found significant associations between lower dietary intakes of folate
and reduced abstract thinking and problem solving performance, but found
no effects for B-12 or B-6. Ortega et al.10, 27 found that those with higher
scores on the Mini Mental State Examination (MMSE) had higher blood lev-
els and dietary intake of folate, but again found no effects for B-12 or B-6.
Other studies have investigated the individual and combined relationships of
folate and B-12 with cognition. Bell et al.28 found that individuals with less
than median values of serum folate and B-12 had lower scores on the MMSE
than those above the median value for either folate or B-12 or both. Wahlin
et al.29 found that those with low serum folate or a combination of low folate
and B-12 performed more poorly on tests of episodic memory than did those
with normal folate and B-12 levels. Hassing et al.30 found that individuals
with low serum folate levels performed more poorly on tests of episodic mem-
ory. Furthermore, there is some evidence to suggest that homocysteine levels
mediate the relationship between the B vitamins and cognitive performance.
Riggs et al.31 found that higher homocysteine, lower folate and B-6 blood lev-
els were associated with poorer spatial copying performance, and that lower
208 THE AGEING BRAIN

B-6 was also associated with backward Digit Span performance. Interestingly,
homocysteine levels were found to mediate the relationships between folate
and B-6 and cognitive performance suggesting that these vitamins impact on
cognition via homocysteine status. The results from these studies examining
cross-sectional associations between B vitamins and cognitive performance
suggest that low folate intake and/or status emerges as the most reliable asso-
ciate of cognitive performance, either alone or in combination with low B-12.
Many aspects of cognition appear to be related to B vitamin status, especially
memory performance and measures of abstract reasoning. In addition, the
relationship between the B vitamins and cognition may be mediated by homo-
cysteine levels since homocysteine uniquely predicted cognitive performance
after controlling for B vitamin status in the study by Riggs et al.31

Longitudinal studies
The findings of cross-sectional studies have been supported by longitudinal
studies. In a six-year follow up of a healthy subsample of the original Good-
win et al.26 study, La Rue et al.32 found a positive association between past
intake of B-12 and B-6 and current cognitive status. Ebly et al.33 found that
those in lowest serum folate quartiles were more likely to have cognitive loss
and to be depressed, and that they were more likely to be demented, institu-
tionalised and to have a higher mortality rate at two-year follow-up. Results
from longitudinal studies allow for the examination of possible long-term
effects of B vitamin intake and status on cognition at a later date, or the
impact on cognitive change. The results of these studies suggest that prior
intake of B vitamins is a predictor of cognitive performance at a later date.
The findings of Ebly et al.33 suggest that low folate status may be a predictor
of cognitive decline among older adults.

Experimental studies
Very few studies have manipulated B vitamin intake experimentally and
assessed its affects on cognitive performance, and only three studies have used
a placebo-controlled design. Tolonen et al.34 investigated B-6 status among
older Finnish and Dutch adults aged between 64 to 96 years, and younger
Dutch adults aged from 22 to 55 years. In addition, they gave daily oral doses
of 2 mg of B-6 for one year to 20 Finnish older adults, while 24 matched par-
ticipants received a placebo. Biochemical results clearly indicated that both
Finnish and Dutch older adults had lower B-6 levels than the younger adults.
Clock drawing performance was improved by supplementation relative to
controls but there were no significant effects of supplementation on memory
or Digit Span performance.
Deijen et al.35 also investigated the effects of B-6 supplementation on cog-
nitive performance and mood among 76 men aged between 70 and 79 years.
They gave the men 20 mg of B-6 or placebo daily for three months. Significant
positive effects of B-6 supplementation, compared with placebo, were found
on measures of the amount of information retained in long-term memory,
NUTRITION AND COGNITIVE AGEING 209

but there were no effects for iconic or short-term memory. The authors con-
cluded that B-6 supplementation might have a modest but significant effect in
improving the storage of information, thereby reducing age-related memory
loss.
Fioravanti et al.36 used a double-blind, placebo-controlled study to assess
the effects of folate supplementation (15 mg daily for 60 days, a dose well
above the Recommended Daily Intake) on the memory performance of 30
community or aged-care dwelling participants, aged 70 to 90 years. Partici-
pants were selected for low folate levels (<3 ng/ml) and mild to moderate
memory complaints. Learning, memory and attention was tested at pre- and
post-supplementation. There were significant differences between the treat-
ment and placebo groups on measures of attention and memory. Moreover,
although there was no significant relationship between folate and cognitive
performance at baseline, the sensitivity of the cognitive measures to treatment
with folate was related to the initial level of the folate deficiency, such that the
greater the deficiency at the start of the treatment, the greater the cognitive
improvement at the end.
In our own laboratory, we recently completed two studies assessing the
effects of short-term folate, B-12 and B-6 supplementation on cognitive per-
formance among healthy, community dwelling women. In the first study,37
a daily capsule of folate (750 micrograms), B-12 (15 micrograms), B-6 (75
milligrams) or placebo was taken for five weeks by 211 women aged 20–92
years. A battery of cognitive tests assessing speed of information processing,
working memory, recall and recognition, executive function and verbal ability
was administered before and after supplementation. Very few effects of supple-
mentation were observed. However, there was a trend for recall performance
among younger women receiving folate and B-12, and among middle-aged
women receiving B-6 to improve from pre- to post-treatment relative to pla-
cebo. Recognition memory performance among older women receiving folate
supplementation improved after treatment relative to placebo. These effects
of supplementation on memory performance were replicated in a follow-up
study38 in which 40 women aged 65 – 84 years received daily doses of 750
micrograms folate and 15 micrograms B-12 or placebo for five weeks. The
immediate and delayed recall performance of those in the vitamin supplement
group improved after treatment while that of the placebo group did not.
Results from well-conducted placebo-controlled experiments would pro-
vide the most compelling evidence of the effects of B vitamins on cognition.
So far, very few have been conducted, but results for the few studies that have
been done suggest that folate, B-12 and B-6 supplementation appear to have
positive effects on the memory performance of older adults. These findings
require replication with an investigation of dose-response relationships.

B vitamins and dementia

There is also an interest in the possible therapeutic role of B vitamins in the
prevention and treatment of Alzheimer’s disease (AD) and other dementias.39
210 THE AGEING BRAIN

Research to date, although equivocal,40 has resulted in recommendations by

the United States National Institute of Aging to conduct a blood chemistry
profile for folate and B-12 and by the Quality Standards Subcommittee of
the American Academy of Neurology41 to screen for B-12 deficiency as part
of the diagnostic process of dementia to rule out preventable causes.42 The
links between the B vitamins, particularly folate and B-12, and dementia are
largely based on findings that individuals with dementia, especially those
with AD, have lower serum folate and B-12 concentrations (3). Furthermore,
serum levels of folate and homocysteine have been found to be related to the
severity of cognitive impairment in dementia.
Several studies have investigated the link between B-12 deficiency and the
incidence of dementia. Karnaze and Carmel 43 analysed serum B-12 levels
in 17 people with primary dementia and 11 with secondary dementia and
found that those with primary dementia (20%) had a higher incidence of
B-12 deficiency than did those with secondary dementia (0%). Ikeda et al.44
also found evidence of B-12 deficiency among those with AD compared to
individuals with other dementias. In this study the deficiency was evident in
cerebrospinal fluid but not plasma concentrations of B-12. In contrast, Crys-
tal et al.45 found that the incidence of dementia did not appear to be related to
B-12 deficiency and that treatment with B-12 did not benefit performance on
measures of cognitive impairment. However, in this study, it was likely that
B-12 deficient participants were present in both the groups with and without
dementia, making differences between the groups difficult to detect.13
Folate status has also been linked with the incidence of dementia, cognitive
impairment and brain atrophy. Sneath et al.46 examined serum folate levels
among 113 patients of a geriatric ward and found that the 14 with dementia
had levels lower than those of the group as a whole. In addition, they found
a positive correlation between red cell folate levels and cognitive perform-
ance scores among those with low folate levels. In support, Sommer and Wol-
kowitz47 reported a positive correlation between red cell folate and scores on
the MMSE among 13 patients with dementia, 10 of whom had a diagnosis
of AD. Most recently, Snowdon et al.25 conducted neuropathological exami-
nations of the brains of 30 participants in the Nun Study who had died and
for whom blood measures were also available. Based on findings that high
homocysteine levels are associated with progressive atrophy of the medial
temporal lobe in those with AD,23 they set out to determine whether serum
folate was inversely related to the severity of atrophy of the neocortex. They
found that serum folate was significantly related to atrophy of the cerebral
cortex but only among those participants with a significant number of AD
lesions. However, since folate was negatively correlated with atrophy even in
participants without brain infarcts and minimal atherosclerosis, the authors
suggest that folate may exert an influence in maintaining CNS integrity in
older age through nonvascular mechanisms.
As discussed earlier, folate and B-12 status may also be more accurately
marked by homocysteine levels and some studies have found associations
NUTRITION AND COGNITIVE AGEING 211

between homocysteine levels and cognitive impairment. Nilsson et al. 48

found increased plasma homocysteine levels in cognitively impaired older
adults with normal blood levels of folate and serum B-12. McCadden et al.49
conducted a prospective case-controlled study of 30 individuals with AD and
found that they had significantly elevated total serum homocysteine levels
compared with a cognitively intact control group. In addition, homocysteine
levels and serum B-12 were correlated with cognitive scores in the AD but
not the control group.
Levitt and Karlinsky50 propose two hypotheses to account for the relation-
ship between B vitamin deficiency and cognitive impairment associated with
dementia. They label the first the “low intake” hypothesis, which proposes
that cognitively impaired individuals have a reduced capacity for self-care
and nutrition and as a consequence develop vitamin deficiency. If this is the
case, then cognitive impairment should be associated with deficiencies in a
wide range of nutritional indices. The second hypothesis is the “etiologic”
hypothesis, which proposes that the B vitamins play a specific role in the
development and severity of cognitive deficits. If this is the case, then the
relationship between B vitamins and cognitive performance should exist for
all types of dementias.
Levitt and Karlinsky50 aimed to test these hypotheses by examining the
relationship between indices of nutrition status and the severity of cognitive
impairment among people with AD and those with other forms of cognitive
impairment. Serum B-12 concentration correlated with MMSE scores only
for the group with AD even after controlling for age, education and duration
of illness. There were no relationships between serum and red cell folate and
cognition scores. The authors concluded that the results did not support the
low intake hypothesis since B-12 was related to cognitive impairment, but
none of the other nutritional indices (folate, magnesium, calcium, protein,
globulin, glucose) was. However, the etiologic hypothesis was not supported
either since the relationship between B-12 and cognition was found only for
those with AD. The authors suggest a third hypothesis which may account
for the association; i.e. the disease process in AD, which results in neuronal
death, may also affect cellular function resulting in a decrease in the absorp-
tion, storage, utilisation and/or excretion of B-12. B-12 levels may therefore
be a marker for the progress of AD. If this is the case, then we might expect
that even though B-12 supplementation may improve the nutritional status
of deficient individuals, it may not impact on cognitive performance. Indeed,
Levitt and Karlinsky50 report that two participants in their study, with low
B-12 and possible AD, who subsequently received B-12 replacement therapy,
continued to show decline in cognitive function despite a return to normal
blood levels of B-12.
The few intervention studies that have been conducted provide some
support for Levitt and Karlinsky’s50 third hypothesis. Teunisse et al.51 gave
B-12 supplementation for six months to 26 individuals with dementia (all
but one had a diagnosis of probable AD) and subnormal B-12 serum levels.
212 THE AGEING BRAIN

The results indicated no effect of supplementation on the severity of cogni-

tive decline among the treated group compared with an untreated group of
individuals with AD. Carmel et al. 52 evaluated B-12, neuropsychological
and electrophysiological indices among 13 individuals with dementia and
low serum B-12 levels, before and after treatment with B-12 supplementa-
tion. Improvements were found for homocysteine and haemoglobin levels,
neuropathological symptoms, electroencephalographic abnormalities, visual
evoked and somatosensory abnormalities, but not on the performance on
neuropsychological tests. Martin et al.53 set out to examine the effects of B-
12 supplementation on cognitive performance in a group of 18 participants
who had low serum cobalamin and evidence of cognitive dysfunction. The
participants received 1000 micrograms of cyanocobalamin intramuscularly
daily for one week, weekly for one month and then monthly for six months.
Post-supplementation scores from the Mattis Dementia Rating Scale for 11 of
the 18 participants showed improvement. However, only those participants
whose impairment on mental status testing had been in the mild range and
who had been symptomatic for less than one year showed improvement. Most
notably, those who had been symptomatic for less than six months responded
best to supplementation. The authors concluded that age-related cognitive
losses in early B-12 deficiency might be reversible if the supplementation is
initiated early in the process.
The findings from these intervention studies generally support Levitt and
Karlinsky’s50 suggestion that AD may be associated with a decrease in the
ability to utilise vitamin B-12. Studies in which B-12 was given as a supple-
ment to those with cognitive impairment resulted in no improvement in the
performance of cognitive tasks, with the exception of the Martin et al. 53 study
in which those who were in the earliest stages of cognitive decline and whose
impairment was in the mild category improved. Thus, it could be argued that
supplementation might be critical in the early stages of AD and that the win-
dow of opportunity for effective intervention with supplementation might be
time limited, since the structural changes in the CNS occurring later in the dis-
ease process will preclude amelioration of cognition.53 Alternatively, studies
in which SAMe, a metabolite of folate and B-12, was given as a supplement54
resulted in improvements in cognition. This may indicate that the utilisation
of SAMe bypasses some crucial metabolic step that is compromised during
the disease. Clearly, more intervention studies are required to determine if
and how B-12 metabolism may be compromised in those with AD.

Antioxidants

The oxidation model of cognitive ageing proposes that there are cumulative
effects of a lifetime of oxidative damage to neuronal tissue in normal age-
ing.32 Neuronal tissue may be particularly vulnerable to oxidative damage
for a number of reasons. First, there is a high content of polyunsaturated
NUTRITION AND COGNITIVE AGEING 213

fatty acids in brain cell membranes,55,56 the relative proportion of which rises
with increasing age, which renders them increasingly sensitive to oxidative
damage with increasing age.57 Second, the brain contains high concentrations
of iron which catalyses the production of oxygen free radical species.55,56
Third, the brain contains limited levels of protective antioxidant enzymes and
compounds.55 In addition, the efficiency of mitochondrial function decreases
with increasing age and defects in mitochondrial energy metabolism have
been linked with oxidative damage. In particular, impaired electron transport
activity in mitochondria is thought to generate free radicals 57 which may
cause oxidative damage to DNA, proteins and lipid membranes. Furthermore,
oxidative modification of low-density lipoprotein is now seen as contributing
to the process of atherogenesis58–60 and there is a link between cerebrovascu-
lar lesions and cognitive impairments.31,58,61
Oxidation in the brain may impact more on some regions than others.
La Rue et al.32 present evidence that shows that with normal ageing there
is an increase in monoamine oxidase B which is involved in the breakdown
in catecholamines. This results in a decrease in dopamine in the striatum
and norepinephrine in the locus ceruleus, spetum and substantia nigra.62
Age-related cognitive changes are consistent with mild impairment of these
frontal/subcortical systems. Such changes include psychomotor slowing, a
decrease in performance of effortful memory tasks and a reduced flexibility
of thought and action. The oxidation model of cognitive ageing therefore sug-
gests a role for antioxidant nutrients, such as vitamins C, E and beta-carotene,
in reducing the effects of oxidative stress in the brain.

Cross-sectional studies
Studies investigating associations between antioxidants and cognitive per-
formance provide mixed support for a role for dietary antioxidants in the
maintenance of cognitive function or reduced cognitive impairment with
increasing age. Jama et al.62 examined the relation between dietary intake
of vitamins C and E and beta-carotene and cognitive function, assessed by
the MMSE, in a population-based sample of 5182 participants aged 55–95.
Results showed that after controlling for sociodemographic variables, total
energy intake and incidence of cardiovascular disease, there was an asso-
ciation between beta-carotene intake and cognitive function. There was no
relation between vitamin C or E intake and cognitive performance. Ortega et
al.27 also investigated associations between dietary intake of antioxidants and
cognitive performance among 260 cognitively unimpaired participants aged
65–90 years. Those with higher MMSE and Pfeiffer’s Mental Status Exam
scores had higher intakes of fruit, vitamin C and beta-carotene compared
with those with lower scores. Mendelsohn et al.63 studied the use of antioxi-
dant supplements among 1059 rural, noninstitutionalised elderly residents
of Pennsylvania. Current use of nutritional supplements containing vitamins
A, C, E, beta-carotene, zinc or selenium were measured via self-report. Anti-
oxidant use was significantly and positively associated with delayed recall
214 THE AGEING BRAIN

of words and story only, out of an extensive battery of neuropsychological

tests, but this association was no longer significant after controlling for age,
education and sex. The researchers concluded that there was little support
for the hypothesis that antioxidant supplements were associated with cogni-
tive function.
Perrig et al.64 examined the relationship between plasma status of vitamins
C and E, and beta-carotene and memory performance among 442 partici-
pants aged between 65 and 94 years. After controlling for age and education,
plasma levels of vitamin C and beta-carotene were correlated with recognition
and vocabulary performance but not with recall and working memory. Vita-
min E status was not associated with cognitive performance. Berr et al.65 also
investigated associations between blood levels of antioxidants and cognitive
performance among a larger sample of 1,389 community-dwelling volunteers
aged 59 to 71 years. Lower plasma carotenoid status was associated with
poorer performance on part B of the Trail Making test and the Digit Symbol
Substitution Test, both of which tap processing speed, but not with memory
performance. As with the findings from Perrig et al.,64 vitamin E status was
not associated with cognitive performance. Perkins et al.66 investigated asso-
ciations between serum levels of a wide range of antioxidants (vitamins A,
C, E, alpha-carotene, beta-carotene, beta-cryptoxanthin, lutein/zeaxanthin,
lycopene and selenium) and delayed recall performance among 4,809 commu-
nity-dwelling adults aged over 60 years. After controlling for age, education,
income, vascular risk factors, and folate, calcium and iron status, lower serum
vitamin E status was associated with a higher odds ratio of having memory
difficulties. There were no other significant associations between blood levels
of any other antioxidant and memory performance.

Longitudinal studies
The results from longitudinal studies examining associations between antioxi-
dants and cognitive performance among older adults provide stronger sup-
port for the link than do cross-sectional results. La Rue et al.32 re-examined
data collected by Goodwin et al.26 and found that plasma concentrations of
vitamin C and use of vitamin C supplements were related to copying, but
not recall, of Rey figure among 137 elderly community residents aged 66–90
years. Other cognitive functions (abstract reasoning, verbal and non-verbal
memory) were not related. Associations between cognition and dietary and
supplementary intake of vitamins A and E were stronger for past (1980) than
current (1986) intake. The researchers suggested that vitamin A stored in the
liver and vitamin E stored in adipose tissue may be available for antioxidant
function at a later time, or antioxidant protection provided by these vitamins
(localized in the lipid-soluble environments in the brain) may prevent slow
oxidative changes that would manifest as poorer cognition later, may account
for these findings. Paleogos et al. 67 conducted a small (N = 117) cohort
study in a retirement community in Sydney, Australia. Vitamin C intake
was assessed in 1991 with a semiquantitative food frequency questionnaire
NUTRITION AND COGNITIVE AGEING 215

with cognitive function assessed 4 years later. After adjustment for age, sex,
smoking, education, total energy intake, and use of psychotropic medica-
tions, higher consumption of vitamin C supplements was associated with
lower prevalence of more severe cognitive impairment measured by MMSE.
However, there were no associations between vitamin C intake and the per-
formance on tests of verbal or category fluency.

Antioxidants and Alzheimer’s disease

Recently there has been an emerging interest in the role of antioxidants in the
treatment of AD. Oxidative damage of cellular lipid, protein and DNA due
to an increase in β-amyloid formation may be central to the neurodegenera-
tive process of AD.68 The hippocampal neurons (the loss of which is a key
feature in AD) seem particularly vulnerable to the deleterious effects of free
radicals on the muscarinic cholinergic receptors.69 Therefore antioxidants,
which tend to act as free-radical scavengers, may have a role in delaying the
course or prevention of AD.
Vitamin E in particular has been the focus of research into the role of
antioxidants in AD. There is some evidence that vitamin E status is altered in
AD. Some studies have found a decrease in plasma concentrations of vitamin
E among those with AD70 while another found increased concentrations of
vitamin E in the brains of those with AD,71 which is possibly indicative of a
compensatory response to oxidative damage. Animal studies using dogs and
rats have demonstrated that an increase in vitamin E concentrations in the
brain can be achieved with supplementation.68 However, to date there has
been very few studies that have investigated the effects of vitamin E on the
course of AD. One such study72 investigated the effects of vitamin E and/or
selegiline supplementation among 341 people with moderate AD using a
double-blind, placebo-controlled design. The risk of institutionalisation, loss
of basic activities of daily living, severe dementia, or death was significantly
reduced among those who received vitamin E and/or selegiline compared with
placebo. There was no effect of supplementation on cognitive test perform-
ance but this could have been due to floor effects or other methodological
problems. This study suggests a possible role for vitamin E and other anti-
oxidants in the prevention or treatment of AD.

Polyunsaturated Fatty Acids

There has been recent interest in the role of omega-3 polyunsaturated fatty
acids (PUFAs) in brain function. The brain is extremely rich in PUFAs, espe-
cially docosahexaenoic acid (DHA; omega-3) and arachidonic acid (AA;
omega-6), which play a critical role in the regulation of membrane perme-
ability and fluidity, as well as in the actions of membrane-bound enzymes
and neurotransmitter (dopamine, serotonin) mechanisms. 73 Fatty fish and
flaxseed oil are major dietary sources of omega-3 PUFAs, and foods high
216 THE AGEING BRAIN

in omega-6 PUFAs include vegetable oils, lean meats, margarines, and eggs.
Upon consumption, the PUFAs compete metabolically in their utilization
of the enzyme delta-6-desaturase to produce their longer-chain derivatives.
Therefore, a diet richer in one PUFA and poorer in the other will result in a
net imbalance within the membrane structure. Typically, the Western diet is
high in omega-6 PUFAs and low in omega-3 PUFAs.74 In addition, increased
levels of omega-6 have been associated with increased production of precur-
sors to cardiovascular disease,75 whereas increasing omega-3 consumption
has been found to attenuate production of these precursors.76 Specifically,
omega-3 PUFAs may down-regulate omega-6 eicosanoids, which tend to be
prothrombotic, vasoconstrictory and proinflammatory, as well as reducing
blood viscosity and improving arterial compliance.75 These effects may have
implications for the oxygen supply to the brain thereby affecting cognitive
function.
Further, the net imbalance of omega-3:omega-6 ratio, or omega-3 defi-
ciency, has been associated with psychological consequences. Omega-3
PUFA deficiency has been linked with changes in cortical dopamine func-
tion77 that may have implications for cognitive functions associated with
the frontal lobes of the brain. Rats fed an omega-3 deficient diet were found
to be impaired on a working memory sensitive version of the Morris water
maze.78 Findings from other animal studies suggest that omega-3 PUFAs may
be important for cognitive function with ageing. A decrease in both DHA and
AA has been found in the brain lipids of older rats and it has been proposed
that these changes in fatty acid composition are correlated with an age-related
decline in the functions of the CNS.79 Further, Lim and Suzuki80 found that
diets rich in DHA and phosphatidylcholine improved maze learning ability in
both younger and older mice. In humans, there have been some studies assess-
ing the importance of cognitive development in infants,81 but there is limited
work on the effects of PUFA intake on cognitive performance among adults.
To date there has been only one investigation,58 that found a link between
fish consumption and rate of cognitive impairment across a three-year time
span among older adults, with higher fish consumption relating to lower rates
of decline, assessed solely by the MMSE. Clearly, further work investigating
the role of PUFAs in cognitive ageing is warranted.

Herbal Supplements — Ginkgo biloba

Lately, attention has been focussed on the potential role of herbal supplements,
such as Ginkgo biloba, in the enhancement of cognitive function. Use of
such supplements, freely available in “health” shops and supermarkets,
may be particularly attractive to older adults seeking to improve cognitive
performance. The therapeutic benefits of Ginkgo biloba are thought to be
due to the action of flavonoids (ginkgo-flavone glycosides) and/or terpenoids
(ginkgolides and bilobalide) which are contained in the extract produced
NUTRITION AND COGNITIVE AGEING 217

from the dried leaves of the plant (Ginkgo biloba, maidenhair tree). Accord-
ing to a review by Kleijnen and Knipschild,82 Ginkgo biloba is among the
most commonly prescribed drugs in France and Germany and is thought to
relieve many neurological conditions common in ageing through a number
of potential mechanisms: increased blood flow by the vasoregulating activ-
ity of arteries, capillaries and veins; platelet activating factor antagonism by
ginkgolides which improve cerebral metabolism and protect the brain against
hypoxic damage; metabolic changes as demonstrated by a reduction in EEG
theta proportion among older adults; and prevention of cell membrane dam-
age caused by free radicals due to the antioxidant properties of ginkgo fla-
vonoids. Ginkgo biloba is thought to help poor concentration or memory,
absent-mindedness, confusion, lack of energy, tiredness, decreased physical
performance, depressive mood, anxiety, dizziness, tinnitus and headache.
These deficits and symptoms are thought to be associated with impaired cer-
ebral circulation and may be early indicators of dementia.82
Of a total of over forty published studies on the efficacy of Ginkgo biloba,
only four84–87 have used standardised neuropsychological tests of cognitive
performance. Most of the studies have used subjective reports of cognitive
performance from participants, doctors or caregivers and most have employed
clinical, elderly samples. Even so, findings from this research suggest that
Ginkgo biloba may be important for optimal cognitive function in older
adults. Only one study has found no effect of Ginkgo biloba on cognition.85
The findings from this study may be compelling because much effort was
made to produce a placebo that could not be distinguished in terms of after-
taste from the Ginkgo biloba supplement. However another interpretation
of these findings was that the supplement group contained a heterogeneous
group of participants in terms of cognitive impairment which may have served
to weaken any effects. Further research on the therapeutic effects and mecha-
nisms by which Ginkgo biloba acts on cognitive performance is needed.

Summary

There is supporting evidence that dietary intake and nutrients can indeed
affect cognitive performance of older adults. Well-conducted research is
emerging and there is a growing interest in important methodological consid-
erations.1 So far most of the conclusions are based on cross-sectional studies
and there have been very few dietary intervention studies using randomised,
placebo-controlled designs. Studies employing such designs are time-consum-
ing and expensive but are essential if we are to discover the mechanisms by
which food components impact on cognitive performance. Further, the effects
of nutrition on cognition are likely to be subtle and complex, with small
effect sizes. Therefore future research needs to be guided by clear hypotheses
about the possible mechanisms by which specific nutrients might affect the
brain and thus cognitive performance so that pertinent and sensitive cogni-
218 THE AGEING BRAIN

tive outcome measures can be selected. The commonly used tests of cognitive
impairment, such as the MMSE, may not be suitable for detecting the effects
of nutrition among healthy, unimpaired older adults due to ceiling effects and
consequent low variability in performance. Adequate sample sizes are also an
important consideration due to the likely subtle effects of nutrition on cogni-
tive performance. Research on the role of nutrition in cognitive ageing is in its
infancy, but there is growing evidence that nutrition may be an important fac-
tor in altering the course and even preventing age-related cognitive decline.

References

1. Bellisle F, Blundell JE, Dye L, Fantino M, Fern E, Fletcher RJ, Lambert J, Rob-
erfroid M, Specter S, Westenhofer J, Westerterp-Plantenga MS. Functional food
science and behaviour and psychological functions. Br J Nutr.1998; 80 (Suppl 1):
S173–193.
2. Riedel WJ, Jorissen BL. Nutrients, age and cognitive function. Curr Opin Clin
Nutr. 1998; 1:579–5865.
3. Selhub J, Bagley LC, Miller J, Rosenberg IH. B vitamins, homocysteine, and
neurocognitive function in the elderly. Am J Clin Nutr. 2000; 71 (Suppl): 614S–
620S.
4. Bryan J. Cognitive function and its links with nutrition. Proc Nutr Soc Aust.
1998; 22:211–215.
5. Salthouse TA. Theoretical perspectives on cognitive aging. Hillsdale NJ: Erlbaum,
1991.
6. Calvaresi E, Bryan J. B vitamins, cognition and ageing: A review. J Gerontol
Psychol Sci. 2001
7. Rosenberg IH, Miller JW. Nutritional factors in physical and cognitive functions
of elderly people. Am J Clin Nutr. 1992; 55 (Suppl 6); 1237S–1243S.
8. Mazza G. Functional foods: Biochemical and processing aspects. Lancaster PA:
Technomic, 1998.
9. Joosten E, van den Berg A, Riezler R, Naurath JJ, Lindenbaum J, Stabler SP, Allen
RH. Metabolic evidence that deficiencies of vitamin B-12 (cobalamin), folate
and vitamin B-6 occur commonly in elderly people. Am J Clin Nutr. 1993; 58:
468–476.
10. Ortega RM, Manas LR, Andres P, Gaspar MJ, Agudo RR, Jiminez A, Pascual
T. Functional and psychic deterioration in elderly people may be aggravated by
folate deficiency. J Nutr. 1996; 126:1192-1199.
11. Parnetti L, Bottiglieri T, Lowenthal D. Role of homocysteine in age-related vas-
cular and non-vascular diseases. Aging Clin Exp Res. 1997; 9:241-257.
12. Sauberlich HE. Relationship of vitamin B-6, vitamin B-12, and folate to neuro-
logical and neuropsychiatric disorders. In: Bendich A, Butterworth CE editors.
Micronutrients in health and in disease prevention. New York: Marcel Dekker,
1991; 187–218
13. Stabler SP, Lindenbaum J, Allen RH. Vitamin B-12 deficiency in the elderly: cur-
rent dilemmas. Am J Clin Nutr. 1997; 66:741-749.
14. Lindeman RD, Romero LJ, Koehler KM, Liang HC, LaRue A, Baumgartner RN,
Garry PJ. Serum vitamin B12, C and folate concentrations in the New Mexico
Elder Health Survey: Correlations with cognitive and affective functions. J Am
Coll Nutr. 2000; 19:68-76.
15. Hankey GJ, Eikelboom JW. Homocysteine and vascular disease. Lancet. 1999;
354:407–413.
NUTRITION AND COGNITIVE AGEING 219

16. Bottiglieri T, Crellin RF, Reynolds EH. Folate and neuropsychiatry. In: Bailey,
KB, editor. Folate in health and disease. New York: Marcel Dekker, 1995;
435–462
17. Alpert JE, Fava M. Nutrition and depression: The role of folate. Nutr Rev. 1997;
55:145–149.
18. Bottiglieri T. Folate, vitamin B12, and neuropsychiatric disorders. Nutr Rev.
1996; 54:382–390.
19. Fenech M, Aitken C, Rinaldi J. Folate, vitamin B12, homocysteine status and DNA
damage in young Australian adults. Carcinogenesis. 1998; 19:1163–1171.
20. Pancharuniti N, Lewis CA, Sauberlick HE, Perkins LL, Go, RCP, Alvarez JO,
Macaluso M, Acton RT, Copeland RB, Cousins AL, Gore TB, Cornwell PE, Rose-
man JM. Plasma homocysteine, folate, and vitamin B-12 concentrations and risk
for early-onset coronary artery disease. Am J Clin Nutr. 1994; 59:940–948.
21. Selhub J, Jacques PF, Bostom AG, D’Agostino RN, Wilson PWF, Belanger AJ,
O’Leary DH, Wolf PA, Shcaefer EJ, Rosenberg IH. Association between plasma
homocysteine concentrations and extracranial carotid-artery stenosis. New Eng
J Med. 1995; 331:286–291.
22. Ueland PM, Refsum H. Plasma homocysteine, a risk factor for vascular disease:
Plasma levels in health, disease, and drug therapy. J Lab Clin Med. 1989; 114:
473–499.
23. Clark R, Smith AD, Jobst KA, Refsum H, Sutton L, Ueland PM. Folate, vitamin
B12 and serum total homocysteine levels in confirmed Alzheimer disease. Arch
Neurol. 1998; 55:1449–1455.
24. Homocysteine Lowering Triallists’ Collaboration. Lowering blood homocysteine
with folic acid based supplements: Meta-analysis of randomised trials. Brit Med
J. 1998; 346:894–898.
25. Snowdon DA, Tully CL, Smith CD, Riley KP, Markesbery WR. Serum folate and
the severity of atrophy of the neocortex in Alzheimer disease: Findings from the
Nun Study. Am J Clin Nutr. 2000; 71:993–998.
26. Goodwin JS, Goodwin JM, Garry PJ. Association between nutritional status and
cognitive functioning in a healthy elderly population. J Am Med Assoc. 1983;
249:2917-2921.
27. Ortega RM, Requejo AM, Andres P, Lopez-Sobaler AM, Quintas ME, Redondo
MR, Navaia B, Rivas T. Dietary intake and cognitive function in a group of
elderly people. Am J Clin Nutr. 1997; 66:803–809.
28. Bell IR, Edman JS, Marby DW, Satlin A, Dreier T, Liptzin B, Cole JO. Vitamin
B12 and folate status in acute geropsychiatric inpatients: Affective and cognitive
characteristics of a vitamin nondeficient population. Biol Psychiat. 1990; 27:
125–137.
29. Wahlin A, Hill RD, Winblad B, Bäckman L. Effects of serum vitamin B12 and
folate status on episodic memory performance in very old age: A population based
study. Psychol Aging. 1996; 11:487–496.
30. Hassing L, Wahlin A, Winblad D, Bäckman L. Further evidence on the effects
of vitamin B12 and folate levels on episodic memory functioning: A population-
based study on healthy very old adults. Biol Psychiat. 1999; 45:1472–1480.
31. Riggs KM, Spiro A, Tucker K, Rush D. Relations of vitamin B-12, folate, and
homocysteine to cognitive performance in the Normative Aging Study. Am J Clin
Nutr. 1996; 63:306–314.
32. La Rue A, Koehler KM, Wayne SJ, Chiulli SJ, Haaland KY, Garry PJ. Nutritional
status and cognitive functioning in a normally aging sample: A 6-year reassess-
ment. Am J Clin Nutr. 1997; 65:20–29.
33. Ebly EM, Schaefer JP, Campbell NRC, Hogan DB. Folate status, vascular disease
and cognition in elderly Canadians. Age Ageing. 1998; 27:485–491.
220 THE AGEING BRAIN

34. Tolonen M, Schrijver J, Westermarck T, Halme M, Touminen SEJ, Frilander A,

Keinonen M, Sarna S. Vitamin B-6 status of Finish elderly. Comparison with
Dutch younger adults and elderly. The effect of supplementation. Int J Vitam
Nutr Res. 1988; 58:73–77.
35. Deijen JB, van der Beek EJ, Orlebeke JF, van den Berg H. Vitamin B-6 supplemen-
tation in elderly men: effects on mood, memory, performance and mental effort.
Psychopharmacology. 1992; 109:489–496.
36. Fioravanti M, Ferrario E, Massaia M, Cappa G, Rivolta G, Grossi E, Buckley
AE. Low folate levels in the cognitive decline of elderly patients and the efficacy
of folate as a treatment for improving memory deficits. Arch Gerontol Geriat.
1997; 26:1–13.
37. Bryan J, Calvaresi E, Hughes D. The effects of short-term folate, B-12 and B-6
supplementation and dietary intake on cognition and mood in women. J Nutr.
(under revision)
38. Rundle B. The effect of vitamin B-12 and folate on cognitive performance and
psychological well-being. Unpublished Honours thesis. University of Adelaide,
2000.
39. Nourhashémi F, Gillet-Guyonnet S, Andrieu S, Ghisolfi A, Ousset PJ, Grandjean
H, Grand A, Pous J, Vellas B, Albarede JL. Alzheimer’s disease: protective factors.
Am J Clin Nutr. 2000; 71 (Suppl):643S–649S.
40. Piccini C, Bracco L, Amaducci L. Treatable and reversible dementias: and update.
J Neurol Sci. 1998; 153:172–181.
41. Knopman DS, De Kosky ST, Cummings JL, Chui H, Corey-Bloom J, Relkin N,
Small GW, Miller B, Stevens JC. Practice parameter: Diagnosis of dementia (an
evidence-based review). Report of the Quality Standards Subcommittee of the
American Academy of Neurology. Neurology. 2001; 56:1143–1153.
42. Pary R, Tobias CR, Lippmann S. Dementia: What to do. Southern Med J. 1990;
83:1182–1189.
43. Karnaze DS, Carmel R. Low serum cobalamin levels in primary degenerative
dementia. Arch Int Med. 1987; 147:429–431.
44. Ikeda T, Furukawa Y, Mashimoto S, Takahashi K, Yamada M. Vitamin B 12
levels in serum and cerebrospinal fluid of people with Alzheimer’s disease. Acta
Psychiat Scand. 1990; 82:327–329.
45. Crystal HA, Ortof E, Frishman WH. Serum vitamin B 12 levels and incidence
of dementia in a health elderly population: a report from the Bronx longitudinal
aging study. J Am Geriatric Soc. 1994; 42:933–936.
46. Sneath P, Chanarin I, Hodkinson HM, McPherson CK, Reynolds EH. Folate
status in a geriatric population and its relation to dementia. Age Ageing. 1973;
2:177–182.
47. Sommer BR, Wolkowitz OM. RBC folic acid levels and cognitive performance in
elderly patients: a preliminary report. Biol Psychiat. 1988; 24:352–354.
48. Nilsson K, Gustafson L, Faldt R, Andersson A, Hultberg B. Plasma homocysteine
in relation to serum cobalamin and blood folate in a psychogeriatric population.
Eur J Clin Invest. 1994; 24:600–606.
49. McCaddon A, Davies G, Hudson P, Tandy S, Cattell H. Total serum homocysteine
in senile dementia of Alzheimer type. Int J Geriatr Psych. 1998; 13:235–239.
50. Levitt AJ, Karlinsky H. Folate, vitamin B 12 and cognitive impairment in patients
with Alzheimer’s disease. Acta Psychiat Scand. 1992; 86:301–305.
51. Teunisse AEB, von Gool WA, Walstra GJM. Dementia and subnormal levels of
vitamin B 12: effects of replacement therapy on dementia. J Neurol. 1996; 243:
522–529.
52. Carmel R, Gott PS, Waters CH. The frequently low cobalamin levels in dementia
usually signify treatable metabolic, neurologic and electrophysiologic abnormali-
ties. Eur J Haematol. 1995; 54:245–253.
NUTRITION AND COGNITIVE AGEING 221

53. Martin DC, Francis J, Protetch J, Huff FJ. Time dependency of cognitive recovery
with cobalamin replacement: Report of a pilot study. J Am Geront Soc. 1992; 40:
168–172.
54. Fontanari D, Di Plama C, Giorgetti F, Violante F, Voltolina M. Effects of S-
Adenosyl-l-methionine on cognitive and vigilance functions in the elderly. Current
Therapeutic Res. 1994; 55:682–689.
55. Carney JM, Starke-Reed PE, Oliver CN, Landum RW, Cheng MS, Wu JF, Floyd
RA. Reversal of age-related increase in brain protein oxidation, decrease in
enzyme activity, and loss in temporal and spatial memory by chronic administra-
tion of the spin-trapping compound N-ert-butyl-α-phenylnitrone. Proc Natl Acad
Sci USA. 1991; 88:3633–3636.
56. O’Donnell E, Lynch MA. Dietary antioxidant supplementation reverses age-
related neuronal changes. Neurobiol Aging. 1998; 19:461-467.
57. Cassarino DS, Bennett JP. An evaluation of the role of mitochondria in neurode-
generative diseases: Mitochondrial mutations and oxidative pathology, protective
nuclear responses, and cell death neurodegeneration. Brain Res Rev. 1999; 29:
1–25.
58. Kalmijn S, Feskens EJM, Launer LJ, Kromhout D. Polyunsaturated fatty acids,
antioxidants, and cognitive function in very old men. Am J Epidemiol. 1997; 145:
33–41.
59. Sack MN, Rader DJ, Cannon RO. Oestrogen and inhibition of oxidation of low-
density lipoproteins in postmenopausal women. Lancet. 1994; 343:269-270.
59. Witztum JL. The oxidation hypothesis of atherosclerosis. Lancet. 1994; 244:
793–795.
60. Rosenberg RN. The aging brain: Limitations in our knowledge and future
approaches. Arch Neurol. 1997; 54:1201–1205.
61. Rogers J, Bloom FE. Neurotransmitter metabolism and function in the aging
nervous system. In Finch CE, Schneider EL. editors. Handbook of the biology of
ageing. New York: Van Nostrand Reinhold, 1995; 645–691.
62. Jama JW, Launer LJ, Witteman JCM, Breeijen JH, Grobbee DE, Hofman A.
Dietary antioxidants and cognitive function in a population-based sample of older
persons. Am J Epidemiol. 1996; 144:275–280.
63. Mendelsohn AB, Belle SH, Stoeher GP, Ganguli M. Use of antioxidant supple-
ments and its association with cognitive function in a rural elderly cohort. Am J
Epidemiol. 1998; 148:38–44.
64. Perrig WJ, Perrig P, Stahelin HB. The relation between antioxidants and memory
performance in the old and very old. J Am Geriatr Soc. 1997; 45:718–724.
65. Berr C, Richard MJ, Roussel AM, Bonithon-Kopp C. Systemic oxidative stress
and cognitive performance in the population-based EVA study. Free Radical Biol
Med. 1998; 24:1202–1208.
66. Perkins AJ, Hendrie HC, Callahan CM, Gao S, Unverzagt FW, Xu Y, Hall KS,
Hui SL. Association of antioxidants with memory in a multiethnic elderly sample
using the Third National Health and Nutrition Examination Survey. Am J Epide-
miol. 1999; 150:37–44.
67. Paleogos M, Cumming RG, Lazarus R. Cohort study of vitamin C intake and
cognitive impairment. Am J Epidemiol. 1998; 148:45–50.
68. Grundman M. Vitamin E and Alzheimer’s disease: the basis for additional clinical
trials. Am J Clin Nutr. 2000; 71(Suppl): 630S–636S.
69. Lethem R, Orrell M. Antioxidants and dementia. Lancet. 1997; 349:1189–
1190.
70. Zaman Z, Roche S, Fielden P, Frost PG, Niriella DC, Cayley AC. Plasma concen-
trations of vitamins A and E and carotenoids in Alzheimer’s disease. Age Aging.
1992; 21:91–94.
222 THE AGEING BRAIN

71. Adams JD, Klaidman LK, Odunze IN, Shen HC, Miller CA. Alzheimer’s and
Parkinson’s disease. Brain levels of glutathione, glutathione disulfide, and vitamin
E. Mol Chem Neuropathology. 1991; 14:213–226.
72. Sano M, Ernesto C, Thomas RG. A controlled trial of selegiline, alpha-toco-
pherol, or both as a treatment for Alzheimer’s disease. The Alzheimer’s Disease
Cooperative Study. N Eng J Med. 1997; 336:1216–1222.
73. Bruinsma KA, Taren DL. Dieting, essential fatty acid intake, and depression. Nutr
Rev. 2000; 58:98–108.
74. Stoll AL, Locke CA, Marangell LB, Severus, WE. Omega-3 fatty acids and bipolar
disorder: A review. Prostag Leukotr Ess. 1999; 60:329–337.
75. Kinsella JE, Lokesh B, Stone, RA. Dietary n-3 polyunsaturated fatty acids and
amelioration of cardiovascular disease: Possible mechanisms. Am J Clin Nutr.
1990; 52:1–28.
76. Endres S, Ghorbani R, Kelley VE, Georgilis K, Lonnemann G, van der Meer
JWM, Cannon JG, Rogers TS, Klempner MS, Weber PC, Schaefer EJ, Wolfe SM,
Dinarello. The effect of dietary supplementation with n-3 polyunsaturated fatty
acids on the synthesis of interleukin-1 and tumor necrosis factor by mononuclear
cells. N Eng J Med. 1989; 320:265–271.
77. Delion S, Chalon S, Hearault J, Guilloteau D, Besnard JC, Durannd, G. Chronic
dietary-linolenic acid deficiency alters dopaminergic and serotoninergic neuro-
transmission in rats. J Nutr. 1994; 124:2466–2476.
78. Wainwright PE, Xing HC, Girard T, Parker L, Ward, GR. Effects of dietary (n-
3) fatty acid deficiency on Morris water-maze performance and amphetamine-
induced conditioned place preference in rats. Nutr Neurosci. 1998; 1:281–293.
79. Söderberg M, Edlund C, Kristensson K, Dallner G. Fatty acid composition of
brain phospholipids in aging and Alzheimer’s disease. Lipids. 1991; 26:421–
425.
80. Lim SY, Suzuki H. Intakes of dietary docosahexaenoic acid ethyl ester and egg
phophatidylcholine improve maze-learning ability in young and old mice. J Nutr.
2000; 130:1629–1632.
81. Wainwright PE. Invited commentary: Nutrition and behavior: The role of n-3
fatty acids in cognitive function. Br J Nutr. 2000; 83:337–339.
82. Kleijnen J, Knipschild P. Gingko biloba. Lancet. 1992; 340:1136–1139.
83. Allaine H, Raoul P, Lieury A, LeCoz F, Gandon JM, d’Arbigny P. Effect of two
doses of Ginkgo biloba extract (Egb 76) on the dual-coding test in elderly sub-
jects. Clinical Therapeutics: Int J Drug Ther. 1993; 15:549–558.
84. Stough C, Clarke J, Lloyd J, Nathan PJ. Neuropsychological changes after 30 day
ginkgo biloba administration in healthy participants. Int J Neuropsychoph. 2001;
4:131–134.
85. van Dongen MJM, van Rossum E, Kessels AGH, Sielhorst HJG, Knipschild PG.
The efficacy of Ginkgo for elderly people with dementia and age-associated
memory impairment: New results of a randomized clinical trial. J Am Geriatr
Soc. 2000; 48:1183–1194.
86. Wesnes K, Simmons D, Rook M, Simpson P. A double-blind placebo-controlled
trial of Tanakan in the treatment of idiopathic cognitive impairment in the eld-
erly. Hum Psychopharm. 1987; 2:159–169.
Chapter 13

THE BRAIN RESERVE

HYPOTHESIS
Peter W Schofield*

Introduction

Anatomical and functional changes are fundamental to the process of age-

ing. In the brain, as elsewhere in the body, changes may reflect ageing alone
or the consequences of age-related diseases.1 The relationship between brain
structure and function and the factors that may influence that relationship
are the principal concerns of this chapter. “Brain reserve” evolved as a model
that helped to explain some otherwise puzzling disjunctions between brain
pathology and brain function. My goal in this chapter is to critically review
the concept of brain reserve, particularly with respect to cognition and
dementia in the elderly.

Background and Definitions

In 1937, Rothschild2 drew attention to the “the inconsistency between clini-

cal and neuropathological phenomena in the field of senile conditions.” Tom-
linson et al. extended this observation using careful quantitative techniques,
correlating the cognitive performance of elderly individuals just prior to their
death to indices of brain pathology found at autopsy.3,4 Using this approach,
they found that modest brain changes, e.g. senile plaque counts below 15
per low power field, or brain softening due to stroke of up to 50 ml, were
associated with normal premortem cognition. Only when senile plaque count,
or stroke volume, exceeded a certain threshold was cognitive impairment a
consistent premortem finding.

*To whom correspondence should be addressed.

224 THE AGEING BRAIN

The notion that the brain exhibits a degree of redundancy, which permits
limited damage to be tolerated without apparent functional consequence,
came to be expressed in the term reserve.5,6 Reserve has been likened to a
buffer that is progressively eroded by accumulating pathology. Brain reserve
theory has most often been invoked in the setting of progressive dement-
ing diseases, particularly Alzheimer’s disease (AD), although it does have
relevance for other conditions such as Parkinson’s disease, stroke, and HIV
infection.
Some ambiguity exists with regard to the term “reserve”, as it is used in
relation to cognition and brain diseases. For example, some authors invite a
distinction between “brain reserve” and “cognitive reserve”.7 Brain reserve,
it is suggested, represents a model in which some structural characteristic of
the brain affords protection against the functional consequences of damage
or disease. By contrast, “cognitive reserve” is represented as a more dynamic
model of compensation for brain damage for which no structural correlate is
identified; the distinction between the two models is likened to that between
hardware and software. While the distinction may be in part one of scale
— even software has its (micro) structural correlates — the predominant
focus of this chapter will be “brain reserve” as characterised above.
The concept of “threshold” is closely related to that of brain reserve. For
example, Satz, in his comprehensive review of brain reserve theory,6 intro-
duces the term “functional impairment cut-off” to correspond to a threshold
level of preserved brain necessary for intact function. Of course, the dis-
tinction between intact function and dysfunction may be difficult to make
and many cases may fall into a “grey” zone. There may also be conceptual
ambiguities. Thus, according to the hierarchical framework suggested by the
World Health Organization,8 dysfunction may be characterised as “impair-
ment”, “disability” and/or “handicap”. According to this model, poor or
declining cognition would represent an impairment, with or without dis-
ability. Disability is defined as a restricted ability to perform a daily activity
normally. When disadvantaged by impairment or disability, an individual is
regarded as handicapped.
The form and meaning of “brain reserve” may differ in important ways,
depending upon what outcome is chosen as the measure of dysfunction. One
key distinction is that between abnormality and decline, each acceptable as
an index of impairment. Thus, cognitive impairment might be diagnosed on
the basis of a performance more than some specified amount (e.g. two stand-
ard deviations) below accepted norms on cognitive tests. Such an approach
is perhaps more commonly adopted in large epidemiological studies. In
other circumstances — particularly clinic-based studies — cognitive decline,
measured or inferred, forms the basis for identifying cognitive dysfunction
(impairment). Implicit in this latter approach is a greater attention to pre-
morbid cognition. Dementia represents a form of cognitive impairment with
associated disability that has the same two contrasting approaches to diagno-
sis, reflecting a relative emphasis either on the level of cognitive abnormality
THE BRAIN RESERVE HYPOTHESIS 225

Figure 1. Dementia diagnosis: impairment or decline?

Figure 2. Brain reserve: two models.

or on the degree of cognitive decline.9,10 Figure 1 depicts these two differing

approaches to dementia diagnosis.
In parallel with these two contrasting ways of defining cognitive dysfunc-
tion are two distinct concepts of “reserve”. I have characterized them as the
“further to fall” and the “resistance to change” versions,11 and a graphic
depiction of the two differing concepts is represented in Figure 2. As reflected
in Figure 2 panel A, individuals endowed with modest cognitive abilities, just
in excess of criteria scores for cognitive abnormality, are clearly at greater
risk of dipping below the threshold for “cognitive impairment” following a
226 THE AGEING BRAIN

small decline in their performance than are individuals whose baseline per-
formance is superior. In terms of risk of becoming “cognitively impaired”
due to some specific brain lesion, the individual of low pre-morbid cognitive
performance would appear to be considerably disadvantaged: that they have
low “brain reserve” would seem obvious. By contrast, if cognitive decline is
used as the index of cognitive dysfunction Figure 2, panel B, it becomes much
less obvious whether, for example, premorbid cognitive ability would offer
an advantage as far as the impact of a specific brain lesion is concerned.
These particular definitional issues can be set aside in relation to what
might be termed a “weak” version of brain reserve theory, which, I sug-
gest, is non-contentious, but nevertheless worthwhile. The weak version
simply contends that brain reserve is lowered by brain damage, while the
strong version states that brain reserve may also differ systematically among
individuals for reasons other than previously acquired brain damage. Some
of the evidence that has been advanced in support of the strong version
— particularly that which comes from epidemiological studies — can be
reinterpreted using the weak version, so it will be important to consider
the evidence for that first.

Brain Reserve: The Weak Version

While the assertion that brain damage lowers brain reserve may seem self-evi-
dent and trivial, there are several corollaries that are of considerable potential
significance.
First, neurological lesions may be asymptomatic and apparently without
clinical consequence. For example, in studies of patients with atrial fibrilla-
tion or those presenting with a recent ischaemic episode, radiological evidence
of a previous clinically silent stroke has been found in at least 10%, compat-
ible with the notion that limited amounts of damage can be tolerated without
impairment.12,13 Autopsy studies have furnished estimates of the likelihood
that the microscopic changes of AD are present at any age. For virtually all
ages, the proportion of autopsied brains with some senile plaques or neurofi-
brillary tangles far exceeds the proportion of age-matched individuals alive
with clinical disease,14,15 and, clearly, dementia due to AD supervenes after
a protracted “preclinical phase”.16
Second, pre-existing neurological damage, even when asymptomatic, may
exaggerate the clinical impact of a subsequent brain insult. For example, in
a study of head injury, twenty young adults who had previously sustained
a concussion recovered cognitive function more slowly after a second head
injury than did control subjects following their first head injury.17 In stud-
ies of stroke patients, the risk for cognitive impairment and/or the degree
of cognitive impairment increases according to infarct number and infarct
size.18 Among individuals with pre-existing cognitive impairment, there is an
increased risk of delirium following relatively minor stress.19
THE BRAIN RESERVE HYPOTHESIS 227

Third, different conditions may be additive or synergistic in their affect

on brain function. For example, in one recent study, the clinical impact of
AD pathology was significantly accentuated by coexistent vascular disease.
In subjects with subcortical infarcts, fewer neuropathologic lesions of AD
resulted in dementia than in those without infarction.20 A high prevalence of
dementia has been observed in people with learning disabilities even when
not attributable to Down’s syndrome21.
Fourth, ageing itself is associated with loss of brain reserve, even in the
absence of specific age-related diseases, as reflected in the greater functional
impact that well-characterised neurological insults have on the aged com-
pared with younger individuals.22,23
As an aside, it is well to remember that small, critically located lesions may
produce major neurobehavioural syndromes and that not all regions of the
brain contribute equally to the integrity of cognitive function.24

Brain Reserve: The Strong Version

In several early studies, some individuals with apparently intact cognition

prior to their death had abundant brain changes at autopsy.3, 25 Such obser-
vations led to a search for individual characteristics or attributes associated
with a heightened capacity to tolerate brain damage. A number of factors
have been identified, consistent with a strong version of brain reserve theory,
which contends that there are non-pathological correlates of reserve.
In general, the evidence bearing on this question has come from three
types of studies. It is worth emphasising that, independent of the major meth-
odological differences outlined below, the studies differ also in the strategies
employed to detect impairment — in particular either “abnormality” or
“decline” — as discussed earlier.

Clinicpathological studies
Here, investigators have examined the relationship between the extent of
brain pathology at autopsy and premortem cognition, seeking systematic
differences that might be accounted for by certain characteristics of the
patients. One caveat for the interpretation of such studies relates to the
uncertain neuropathological basis for cognitive changes in dementing
diseases. Thus, in the case of AD, the visible pathology of plaques and
tangles have been used as an index of the extent of pathology, while there
is evidence to suggest that synaptic loss may be the proximate cause of
cognitive decline.26

Imaging studies
Brain atrophy or functional brain imaging abnormalities in life can be used to
provide a proxy for pathology and related to cognition, or some other meas-
ure of brain function. Once again, investigators seek systematic variation in
228 THE AGEING BRAIN

the relationship between “pathology” and cognition/brain function according

to some defining characteristic of the subjects.

Epidemiological studies
These offer a means of identifying risk or protective factors for cognitive
impairment, cognitive decline, or dementia. Because the common causes of
dementia — AD, stroke — are strongly associated with age,27 it is plausible
to suggest that protective factors may have their effect by delaying the clinical
onset of disease by virtue of an association with increased brain reserve. It is
possible, of course that “protective” factors may not operate via a mechanism
of “reserve”, but by some other means, as will be discussed below.
From these various investigative approaches, a number of factors have
been suggested as possible correlates of brain reserve.
Low educational attainment has emerged in numerous epidemiological
studies as a risk factor for AD28–40 and in some for the development of
stroke-related dementia.41–44 The same studies can be interpreted to mean
that higher educational attainment is protective. Other studies employing
different methods have added weight to these findings, but not all the rel-
evant studies have been in agreement. Other factors that have been advanced
as correlates of increased brain reserve include higher “intelligence”, 45–47
increased brain size,25,48–51 certain occupational types33,41,52 and increased
mental activity.53–57 I propose to review some of this evidence below.

Education and Occupation

Low educational attainment has been examined as a potential risk factor for
the development of dementia in the context of several common conditions
affecting the elderly including stroke, Parkinson’s disease, and AD.
In one case control study of stroke, subjects with less than eight years
of formal schooling were more than 40 times more likely to have dementia
than other stroke subjects with more education, after adjusting for other risk
factors, including hypertension, recent smoking, obesity, and proteinuria.42
Other studies have obtained more modest estimates of the effect size.43,44
Education has generally not been found to be a risk factor for dementia
associated with Parkinson’s disease although borderline associations have
been reported. In one prospective cohort study of community-dwelling
patients with Parkinson’s disease, subjects who developed dementia during a
3.5 year follow-up period had an average of two years less formal education
than subjects who remained non-demented.58
Numerous studies have reported an association between low educational
attainment and frequency of AD.28–41 In an early, important study, the preva-
lence of dementia (most due to AD) was estimated among individuals over
65 years in Shanghai, China.29 In the initial screening phase of this study,
a cognitive test with education-dependent cutoffs was used and, during the
THE BRAIN RESERVE HYPOTHESIS 229

subsequent diagnostic phase, neuropsychological tests were used that were

relatively invariant to education. Clinical evaluations were performed and a
history of functional decline was required for the diagnosis of dementia. Age,
female gender, and low education were all highly significantly, independently
associated with increased prevalence of AD. The association of AD with
education was particularly striking, in view of the investigators’ efforts to
minimize or eliminate bias with respect to education in the diagnostic proc-
ess.
A study in New York City found that educational attainment and occu-
pational status were both associated with differential risk for incident AD.
Subjects with low education and low occupational status were at greatest
risk for dementia at follow-up.33 Similar findings were reported from a study
conducted in East Boston.36
More recent epidemiological studies have produced new twists. In both
the Rotterdam study,38 and in the EURODEM pooled analysis,59 the asso-
ciation between low education and AD was present in women only. Among
men, there was no association between low educational attainment and risk
of AD. The authors concluded that unmeasured confounding might be the
explanation for previous findings. In another study, low educational attain-
ment was associated with increased risk of dementia among individuals with a
childhood history of rural but not urban residence.35 And in an Italian study,
having no education was associated with increased risk for dementia, but
individuals with little education (less than three years) were no more likely
than better- educated individuals to be demented.37
In support of the hypothesis that individuals with higher educational attain-
ment become demented later in the course of the underlying disease, several
groups have noted that their subsequent mortality may be increased,60,61 or
their decline on neuropsychological tests more rapid,62 relative to comparably
demented individuals with less education.
Several studies have found no association between education and AD. In
a case-control study that used record-linkage data from Rochester, no differ-
ence was found between the educational level of cases with AD and controls,
most of whom had more than nine years of education.63 In a study of incident
dementia and AD from Framingham, low educational attainment was not a
risk factor for AD, although it was a risk factor for non-AD dementia.64 This
study sample also included relatively few individuals with low educational
attainment: more than 60% reported at least a high school education. A
British study also found no differences in incidence of dementia in relation to
educational attainment, after adjustment for age.65
Lower educational attainment has been identified as a risk factor for cogni-
tive decline with ageing,66,67 as well as cognitive impairment (as distinct from
a measured decline) without dementia in a number of report.68,69
In several studies, SPECT scanning has been used to provide an index of
the extent of pathology in patients with AD. Stern and colleagues showed
that, for a given level of cognitive performance on testing, individuals with
230 THE AGEING BRAIN

greater educational attainment had greater flow deficits on SPECT scans.70

This result suggests that, in better-educated individuals with AD, cognition
is preserved longer into the illness than in those with less education. A later
study employing similar methodology provided evidence to suggest that more
intellectually challenging occupations confer similar benefits for individuals
with AD.71
There are few clinicopathological studies relevant to the possible role of
education in early life as a determinant of brain reserve. However, in one
fascinating study, the brains of 20 subjects who had been free of cognitive
disorders prior to death were examined at autopsy for measures of dendritic
complexity, and these measures were correlated with sociodemographic char-
acteristics.72 Educational attainment in life was strongly, positively correlated
with dendritic complexity. This finding is consistent with animal data that
suggests that both neuronal development and connectivity are promoted by a
stimulating environment during early life.73 Of course, it was not possible for
the investigators to determine whether educational attainment caused or was
a consequence of the favourable brain anatomy, but it does at least suggest a
possible structural correlate of reserve.
A recent autopsy study of patients attending a memory disorders clinic
has been advanced as evidence against the brain reserve theory74 at least in
the form I refer to as the “strong version”. In this study, 87 patients who had
been followed in a memory disorders clinic and who had come to autopsy
were compared with respect to sociodemographic, clinical and autopsy data.
Less-educated patients with dementia were on average older at presentation
and had more cerebrovascular pathology than better-educated demented
patients. The authors of this study proposed that the apparent association
between low education and higher frequency of AD suggested by numerous
epidemiological studies might be due to a relative excess of cerebrovascular
disease among less educated individuals. Other investigators have demon-
strated an excess of cardiovascular risk factors75 and risk for stroke76 among
individuals with less educational attainment, and asymptomatic stroke may
be higher in this group as well. It is certainly plausible that lowered brain
reserve, secondary to subclinical stroke-related brain damage, might account
for some of the epidemiological findings outlined above. Future epidemiologi-
cal studies in which participants undergo routine brain imaging will allow
this possibility to be carefully evaluated.
To summarize, there is evidence from many but not all studies indicating
that increased educational attainment is associated with lower frequencies
of dementia, cognitive abnormality without dementia, and cognitive decline.
There is some support for a threshold effect, such that protection is achieved
with relatively little education.
A proposed mechanism for these findings is that education in some
way leads to increased brain reserve, delaying the clinical onset of disease
— and of AD in particular. Katzman77 has suggested that education may
act by enhancing the development synapses, citing the evidence from animal
THE BRAIN RESERVE HYPOTHESIS 231

studies indicating benefits for early brain development of a “stimulating

environment”, referred to above. Because educational attainment also cor-
relates strongly with intellectual ability as measured formally by IQ tests,78
it is possible that greater innate intellectual capacity underlies the findings
reviewed above. Proponents of “cognitive reserve” theory suggest that those
individuals with higher educational attainment or occupational status may
have more cognitive flexibility and a wider repertoire of cognitive responses
which enables them to better tolerate brain damage or disease.7 Alternative
interpretations of the data essentially invoke what I have referred to above
as the “weak form” of brain reserve theory. For instance, it is possible that
limited educational opportunity is a proxy for early exposures to potentially
harmful agents that either impair brain development or more directly lead
to cognitive decline later in life.79 The possibility that clinically silent stroke
may underlie some of the increase in dementia among the poorer-educated
has been touched on already.

Intelligence

The role of premorbid intellect as a determinant of outcome following brain

injury or disease has been examined by a number of investigators. In one
study, Vietnam veterans who had sustained penetrating brain injuries were
assessed on a range of cognitive measures.80 An estimate of pre-injury intel-
ligence was available for all individuals from their performance on cognitive
testing at induction. The volumes of brain tissue loss and lesion location
were significant predictors of post injury cognition but, overall, pre-injury
intelligence and education were the strongest predictors. In another study,
scores on cognitive tests administered to young individuals in the 1940s were
compared with the results of cognitive testing conducted via telephone some
50 years later.81 A strong correlation existed between the two test perform-
ances, indicating that “premorbid” ability in youth is a strong predictor of
cognitive performance at later ages, when the brain is subject to age-related
change or dementing diseases. In a large population-based study of elderly
individuals, Schmand et al. used the Dutch Adult Reading Test (DART) as an
index of premorbid intelligence at base-line assessment and followed initially
non-demented subjects for four years for the development of dementia.45 In
this study, a low performance on the DART at baseline predicted incident
dementia at follow-up better than did a low level of education.
Several studies have cleverly used pre-existing data to provide estimates
of premorbid intelligence. In a well-known study of elderly nuns, an index
of premorbid cognitive ability was obtained by reviewing diary entries that
the nuns (then novices) had made when in their late teens or early twenties.46
The entries were “scored” for linguistic competence and these measures were
related to the risk for subsequent dementia and to the extent of pathological
changes in the brains of those nuns who came to autopsy. Poorer linguistic
232 THE AGEING BRAIN

aptitude was associated with increased risk of dementia, a finding consistent

with reserve theory. An unexpected finding of this study was a positive associ-
ation between poorer linguistic aptitude in youth and the presence of AD-like
brain pathology at autopsy: nuns whose diaries indicated poorer linguistic
performance in youth had on average more brain changes of AD, whether
or not they had clinical dementia. This latter finding is not one that would
be predicted by reserve theory and led the authors to suggest the intriguing
possibility that AD might have its earliest manifestations at a much earlier
age than previously thought. According to this hypothesis, the relatively poor
linguistic performance of nuns destined to develop AD may have represented
its earliest cognitive effects.
In a recent study from Scotland, cognitive ability in early life was also
found to be associated with risk of subsequent dementia.47 Demented and
non-demented individuals, who were all born in 1921, were compared for
their performance on routine cognitive testing conducted in 1932 as part of
the “Scottish Mental Survey”. Compared with controls, demented individu-
als had performed more poorly on the cognitive testing in 1932. The effect
was restricted to cases who had become demented after the age of 64: early
onset dementia cases did not differ from controls with respect to their early
life cognition.
The results from these studies are open to a number of possible interpre-
tations including those deriving from “reserve theory”. Brighter individuals
may manifest enhanced cognitive flexibility with benefits along lines already
suggested in relation to the effects of education and occupation. Alternatively,
premorbid intellect might be associated with greater levels of mental activity
as a more fundamental mechanism of protection, to be addressed in greater
detail below. Intellectually brighter individuals may adopt more healthy
lifestyles, protecting themselves from negative health outcomes including
cerebrovascular disease. Finally, the possibility has been raised that some
of the variability in intellectual capacity in early and middle life reflects the
consequences of pathological processes, AD particularly, for which there is
little currently identifiable neuropathological correlate. According to this
speculation, “premorbid” intellect might represent a consequence of, rather
than an antecedent for, AD.82

Brain Size

In a clinicopathological study of 137 nursing home residents, 10 individu-

als with the neuropathological changes of AD at autopsy had never been
demented.25 These individuals had bigger brains than did the remaining
subjects (either with or without dementia). The authors of this study sug-
gested that the larger brains might have afforded increased reserve, delaying
the onset of clinical disease. In a study of 28 women attending a memory
disorders clinic, each with a diagnosis of probable AD, a cross-sectional intra-
THE BRAIN RESERVE HYPOTHESIS 233

cranial area measurement was obtained from CT brain scans as an index of

premorbid brain size.48 This measure correlated significantly with the age at
onset of symptoms, estimated from informant reports. Individuals with larger
intracranial areas had a later onset of symptoms, consistent with the hypoth-
esis that brain size mediates reserve, delaying onset of clinical features. A
study by Mori and colleagues,50 while differing in important methodological
respects, offered partial support for these findings. A more recent study found
no evidence for an association between intracranial volume (a reliable index
of premorbid brain volume) and age at onset of AD, but this study included
many with early onset disease in which mechanisms of brain reserve may be
relatively unimportant.83
In two published studies, head circumference has been used as a proxy for
premorbid brain size, and evaluated as a risk factor for AD. In a large study of
Japanese Americans, 83 individuals were identified with AD, of whom those
with smaller head circumference were more severely affected, after adjusting
for age and education.49 In a multiethnic, population-based study in New
York City, individuals in the lowest quintile of head circumference for gender
were at increased risk of having AD, compared with those in the upper four
quintiles, even after adjusting for education, age, and ethnicity.51 Height,
weight and APOE genotype were not confounders in this association. Again,
these results are consistent with the notion that brain size mediates reserve,
allowing function to be preserved longer in those with underlying AD.
Interestingly, brain size has been shown to correlate with IQ in methodo-
logically rigorous studies84,85 and superior premorbid intellect may therefore
underlie the apparent protective effect of larger brains, according to mecha-
nisms already discussed. Conversely, it is possible that the protective effects
of higher IQ for AD might be mediated by larger brain size.84,85 The micro-
structural correlates of bigger brains that might account for greater reserve
are uncertain. Increased levels of premorbid synaptic density, neuronal count
or neuronal connectivity might optimise the potential for functional adapta-
tion to neurodegenerative processes. Alternative and highly speculative sug-
gestions are that increases in the glial:neuron ratio or synaptic density may
have neuroprotective consequence.86,87

Mental and other Activity

In a cross-sectional study of nearly 300 elderly volunteers who were

questioned about intellectual activities from which a “total intellectual
activity” score was obtained, this emerged as the second most powerful
positive predictor (after education) of objective memory performance. 88 In
several case control studies, a history of inactivity has been associated with
increased risk of AD. Kondo et al. reported increased risk of AD associated
with a reduction in a variety of “psychosocial behaviours” and “uses of
leisure time” in the fifth and sixth decades of life. 89 In a study by Broe et
234 THE AGEING BRAIN

al., individuals with AD were significantly more likely than controls to be

described as “physically underactive” in the ten years previously, as well as
more than ten years ago.90 In a prospective study, a history of doing odd
jobs, knitting, or gardening was associated with significantly reduced risk
of subsequent dementia in the one to three years of follow-up. 91 In another
longitudinal study, subjects who “participated in novel information process-
ing activities such as learning a language or playing bridge” were less likely
to show cognitive decline during the follow-up period. 92 More recently,
evidence that physical activity may reduce risk for cognitive impairment or
dementia has also been advanced.93
Friedland has suggested that chronic neuronal activation might enhance
neuronal integrity via favorable secondary changes in cerebral blood flow,
calcium homeostasis, amyloid precursor protein turnover, levels of stress, and
DNA repair.94,95 More recently, Gould and others96,97 have demonstrated
in experimental animals that, contrary to longstanding dogma, new neurons
may develop in adult brains. Other investigators have shown that the sur-
vival of these neurons may be enhanced by exposure to an enriched environ-
ment.98,99 These revolutionary findings suggest additional means by which
intellectual activity in humans might lead to enhanced reserve and delay in
the clinical onset of dementing diseases.

Summary and Conclusions

I have stressed that “the brain reserve hypothesis” can be viewed in somewhat
different ways, depending upon varying notions of dysfunction. Independent
of those distinctions, a weak form of brain reserve theory acknowledges the
impact of brain damage, even when this is subclinical, and a strong form of
brain reserve theory proposes that individual characteristics also influence the
functional outcome of brain damage or disease. Evidence in support of both
the weak and strong versions has been reviewed.
What are the implications of the brain reserve hypothesis, as I have pre-
sented it? One key area of concern relates to the interaction between vascular
pathology and AD, with the potential for incipient AD to be “unmasked” by
relatively minor vascular events. The potential benefits of scrupulous atten-
tion to cerebrovascular risk factors are obvious. Individuals who display evi-
dence of “lowered reserve” perhaps by developing confusion in the context
of relatively minor stresses warrant careful evaluation for possible incipient
AD, and in the future might be candidates for empiric treatment with dis-
ease modifying agents when these become available. At a global perspective,
improved infant welfare to optimise brain growth100 and early development
may have benefits in the reduction of cognitive disorders later in life. Educa-
tion may promote brain development, but its benefits in later life might also
reflect the fostering of habits of increased mental activity. Hopefully, educa-
tion also promotes the development of critical faculties leading to avoidance
THE BRAIN RESERVE HYPOTHESIS 235

of unhealthy activities and preferential adoption of healthy ones. Finally,

possible determinants of brain reserve that are under partial genetic control,
such as brain size101 and intelligence,102 might add to the recognised familial
contribution to AD, in so far as they may influence the time at which the
underlying pathology becomes clinically manifest.

References

1. Kemper TL. Neuroanatomical and neuropathological changes during aging and

dementia. In: Albert ML, Knoefel JE, editors. Clinical neurology of aging. 2nd
ed. New York: Oxford University Press, 1994.
2. Rothschild D. The pathological changes in senile psychosis and their
psychobiologic significance. Am J Psychiat. 1937; 93:757.
3. Tomlinson BE, Blessed G, Roth M. Observations on the brains of non-demented
old people. J Neurol Sci. 1968; 7:331–356.
4. Tomlinson BE, Blessed G, Roth M. Observations on the brains of demented old
people. J Neurol Sci. 1970; 11:205–242.
5. Mortimer JA. Do psychosocial risk factors contribute to Alzheimer’s disease? In
Henderson AS, Henderson JH, editors. Etiology of dementia of Alzheimer’s type.
Chichester: John Wiley, 1988; 39–52.
6. Satz P. Brain reserve capacity of symptom onset after brain injury: a formulation
and review of evidence for threshold theory. Neuropsychology. 1993; 3:273–
295.
7. Stern Y. What is cognitive reserve? Theory and research application of the reserve
concept. JINS. 2002; 8:448–460.
8. World Health Organization. WHO international classification of impairments,
disabilities and handicaps: A manual of classification relating to the consequences
of disease. Geneva: World Health Organization, 1980.
9. American Psychiatric Association. Diagnostic and statistical manual of mental
disorders. 4th ed. Washington, DC: American Psychiatric Press Inc,1994.
10. The ICD-10 classification of mental and behavioural disorders: Clinical
descriptions and diagnostic guidelines. Geneva: World Health Organization,
1992.
11. Schofield PW. Alzheimer’s disease and brain reserve. Aust J Ageing. 1999;18:
10–14.
12. Petersen P, Madsen EB, Brun B, Pedersen F, Gyldensted C, Boysen G. Silent
cerebral infarction in chronic atrial fibrillation. Stroke. 1987; 18:1098–1100.
13. Kempster PA, Gerraty RP, Gates PC. Asymptomatic cerebral infarction in patients
with chronic atrial fibrillation. Stroke. 1988; 19:955–957.
14. Morimatsu M, Hirai S, Muramatsu A, Yoshkawa M. Senile degenerative brain
lesions and dementia. J Am Geriatr Soc. 1975; 23:390–406.
15. Ohm T, Muller H, Braak H, Bohl J. Close-meshed prevalence rates of different
stages as a tool to uncover the rate of Alzheimer’s disease-related neurofibrillary
changes. Neuroscience. 1995; 64:209–217.
16. Linn RT, Wolf PA, Bachman DL, et al. The preclinical phase of Alzheimer’s dis-
ease. A 13-year prospective study of the Framingham cohort. Arch Neurol. 1995;
52:485–490
17. Gronwall D, Wrightson P. Cumulative effect of concussion. Lancet. 1974; ii:
995–997
18. Skoog I. Risk factors for vascular dementia: a review. Dementia. 1994; 5:137–
144
236 THE AGEING BRAIN

19. Schor JD, Levkoff SE, Lipsitz LA, Reilly CH, Cleary PD, Rowe, JW, Evans DA.
Risk factors for delirium in hospitalized elderly. J Amer Med Assoc. 1992; 267:
827–831.
20. Snowdon DA, Greiner LH, Mortimer JA, Riley KP, Greiner PA, Markesbery WR.
Brain infarction and the clinical expression of Alzheimer’s disease; the Nun Study.
J Amer Med Assoc. 1997; 277:813–817.
21. Cooper S-A. High prevalence of dementia among people with learning disabilities
not attributable to Down’s syndrome. Psychol Med. 1997; 27:609–616.
22. Mann DMA, Yates PO, Marcyniuk B. Alzheimer’s presenile dementia, senile
dementia of Alzheimer type and Down’s syndrome in middle age form an age
related continuum of pathological changes. Neuropath Appl Neuro. 1984; 10:
185–207.
23. Cryns AG, Gorey KM, Goldstein MZ. Effects of surgery on the mental status of
older persons. A meta-analytic review. J Geriatr Psych Neur. 1990; 3:184–191.
24. Tatemichi TK, Desmond DW, Prohovnik I. Strategic infarcts in vascular demen-
tia. Arznei– Forschung. 1995; 45:371–385.
25. Katzman R, Terry R, DeTeresa R, Brown T, Davies, P, Fulci P, Renbing X, Peck
A. Clinical, pathological, and neurochemical changes in dementia: a subgroup
with preserved mental status and numerous neocortical plaques. Ann Neurol.
1988; 23:138–144.
26. Terry RD, Masliah E, Salmon DP, Butters N, DeTeresa R, Hill R, Hansen LA,
Katzman R. Physical basis of cognitve alterations in Alzheimer’s disease: syn-
apse loss is the major correlate of cognitive impairment. Ann Neurol. 1991; 30:
572–580.
27. Jorm AF, Korten AE, Henderson AS. The prevalence of dementia: a quantitative
integration of the literature. Acta Psychiat Scand. 1987; 76:465–479.
28. Bonaiuto S, Rocca WA, Lippi A, Luciani P, Turtu F, Cavarzeran F, Amaducci
L. Impact of education and occupation on the prevalence of Alzheimer’s disease
(AD) and multi-infarct dementia (MID) in Appignano, Macerata Province, Italy.
Neurology. 1990; 40 (Suppl 1):346.
29. Zhang MY, Katzman R, Salmon D, Jin H, Cai GJ, Wang ZY, Qu GY, Grant I, Yu
E, Levy P. The prevalence of dementia and Alzheimer’s disease (AD) in Shanghai,
China: impact of age, gender and education. Ann Neurol. 1990; 27:428–437.
30. Rocca WA, Bonaiuto S, Lippi A, Luciani P, Turtu F, Cavarzeran F, Amaclucci L.
Prevalence of clinically diagnosed Alzheimer’s disease and other dementing disor-
ders: a door-to-door survey in Appignano, Macerata Province, Italy. Neurology.
1990; 40:626–631.
31. Fratiglioni L, Grut M, Forsell Y, Viitanen M, Grafstrom M, Holmen K, Ericsson
K, Backman L, Ahlbom A, Winbald B. Prevalence of Alzheimer’s disease and
other dementias in an elderly urban population:relationship with age, sex and
education. Neurology. 1991; 41:1886–1892.
32. Korczyn AD, Kahana E, Galper Y. Epidemiology of dementia in Ashkelon Israel.
Neuroepidemiology. 1991; 10:100.
33. Stern Y, Gurland B, Tatemichi T, Tang M, Wilder D, Mayeux R. Influence of
education and occupation on the incidence of Alzheimer’s disease. J Amer Med
Assoc. 1994; 271:1004–1010.
34. Ott A, Breteler M, van Harskamp F, Clauss JJ, van der Cammen TJ, Grobbee DE,
Hofman A. Prevalence of Alzheimer’s disease and vascular dementia: association
with education. The Rotterdam Study. Br Med J. 1995; 310:970–973.
35. Hall KS, Gao S, Unverzagt FW, Hendrie HC. Low education and childhood rural
residence: risk for Alzheimer’s disease in African Americans. Neurology. 2000;54:
95–99.
36. Evans DA, Hebert LE, Beckett LA, Scherr PA, Albert MS, Chown MJ, Pilgrim
DM, Taylor JO. Education and other measures of socioeconomic status and risk
THE BRAIN RESERVE HYPOTHESIS 237

of incident Alzheimer’s disease in a defined population of older persons. Arch

Neurol. 1997; 54:1399–1405.
37. De Ronchi D, Fratiglioni L, Ruci P, Paternico A, Graziani S, Dalmonte E. The
effect of education on dementia occurrence in an Italian population with middle
to high socioeconomic status. Neurology. 1998; 50:1231–1238.
38. Education and the incidence of dementia in a large population-based study: the
Rotterdam Study. Neurology. 1999; 52:663–666.
39. Letenneur L, Gilleron V, Commenges D, Helmer C, Orgogozo JM, Dartigues JF.
Are sex and educational level independent predictors of dementia and Alzheimer’s
disease? Incidence data from the PAQUID project. J Neurol Neurosur Ps. 1999;
66:177–183.
40. Lin R-T, Lai C-L, Liu C-K, Yen Y-Y, Howng S-L. prevalence and subtypes of
dementia in southern Taiwan: impact of age, sex, education, and urbanization. J
Neurol Sci. 1998; 160:67–75.
41. Mortel KF, Meyer JS, Herod B, Thornby J. Education and occupation as risk
factors for dementias of the Alzheimer and ischemic vascular types. Dementia.
1995; 6:55–62.
42. Gorelick PB, Brody J, Cohen D, Freels S, Levy P, Dollear W, Forman H, Harris
Y. Risk factors for dementia associated with multiple cerebral infarcts; a case
control analysis in predominantly African-american Hospital-based patients.
Arch Neurol. 1993; 50:714–720.
43. Tatemichi TK, Desmond DW, Paik M, Figueroa M, Gropen TI, Stern Y, Sano M,
Remien R, Williams JB, Mohr JP. Clinical determinants of dementia related to
stroke. Ann Neurol. 1993; 33:568–575.
44. Desmond DW, Moroney JT, Paik MC, Sano M, Mohr JP, Aboumatar S, Tseng CL,
Chan S, Williams JB, Reiman RH, Hauser WA, Stern Y. Frequency and clinical
determinants of dementia after ischemic stroke. Neurology. 2000; 54: 1124–1131.
45. Schmand B, Smit JH, Geerlins MI, Lindeboom J. The effects of intelligence and
education on the development of dementia. A test of the brain reserve hypothesis.
Psychol Med. 1997; 27:1337–1344.
46. Snowden DA, Kemper SJ, Mortimer JA, Greiner LH, Wekstein DR, Markesbery
WR. Linguistic ability in early life and cognitive function and Alzheimer’s disease
in late life. J Amer Med Assoc. 1996; 275:528–532.
47. Whalley LJ, Starr JM, Athawes R, Hunter D, Pattie A, Deary IJ. Childhood
mental ability and dementia. Neurology. 2000; 55:1455–1459.
48. Schofield PW, Mosesson R, Stern Y, Mayeux R. The age at onset of Alzheimer’s
disease and an intracranial area measurement: a relationship. Arch Neurol. 1995;
52:95–98.
49. Graves AB, Mortimer JA, Larson EB, Wenzlow A, Bower JD, McCormick WC.
Head circumference as a measure of cognitive reserve. Association with severity
of impairment in Alzheimer’s disease. Br J Psychiat. 1996; 169: 86–92.
50. Mori E, Hirono N, Yamashita H, Imamura T, Ikejiri Y, Ikeda M, Kitagaki H,
Shimomura T, Yoneda Y. Premorbid brain size as a determinant of reserve
capacity against intellectual decline in Alzheimer’s disease. Am J Psychiat. 1997;
154:18–24.
51. Schofield PW, Logroscino G,Andrews HF, Albert S, Stern Y. An association
between head circumference and Alzheimer’s disease in a population-based study
of aging and dementia. Neurology. 1997; 49:30–37.
52. Dartigues JF, Gagnon M, Mazaux JM, Barberger-Gateau P, Commenges D,
Letenneur L, Orgogozo JM. Occupation during life and memory performance
in nondemented French elderly community residents. Neurology. 1992; 42:
1697–1701.
53. Arbuckle TY, Gold D, Andres D. Cognitive functioning of older people in relation
to social and personality variables. J Psychol Aging. 1986; 1:55–62.
238 THE AGEING BRAIN

54. Broe GA, Henderson AS, Creasey H, McCusker E, Korten AE, Jorm AF, Longley
W, Anthony JC. A case-control study of Alzheimer’s disease in Australia. Neurol-
ogy. 1990; 40:1698–1707.
55. Kondo K, Niino M, Shido K. A case-control study of Alzheimer’s disease in Japan
– significance of life-styles. Dementia. 1994; 5:314–326.
56. Fabrigoule C, Letenneur L, Dartigues JF, Zarrouk M, Commenges D, Barberger-
Gateau P. Social and leisure activities and risk of dementia: a prospective longi-
tudinal study. J Am Geriatr Soc. 1995; 43:485–490.
57. Gold DP, Andres D, Etezadi J, Arbuckle T, Schwartzman A, Chaikelson J. Struc-
tural equation model of intellectual change and continuity and predictors of
intelligence in older men. Psychol Aging. 1995 ;10:294–303.
58. Marder K, Tang M-X, Cote L, Stern Y, Mayeux R. The frequency and associated
risk factors for dementia in patients with Parkinson’s disease. Arch Neurol. 1995;
52:695–701.
59. Letenneur L, Launer LJ, Andersen K, Dewey ME, Ott A, Copeland JR, Dartigues
JF, Kragh-Sorensen P, Baldereschi M, Brayne C, Lobo A, Martinez-Lage JM,
Stijnen T, Hofman A. Education and the risk for Alzheimer’s disease: sex makes
a difference. Am J Epidemiol. 2000; 151:1064–1071.
60. Stern Y, Tang M-X, Denaro J, Mayeux R. Increased risk of mortality in Alzheim-
er’s disease with more advanced educational and occupational attainment. Ann
Neurol. 1995; 37:590–595.
61. Geerlings MI, Deeg DJH, Pennix BWJH, Schmand B, Jonter C, Bouter LM, van
Tilburg W. Cognitive reserve and mortality in dementia: the role of cognition,
functional ability and depression. Psychol Med. 1999; 29:1219–1226.
62. Stern Y, Albert S, Tang M-X, Tsai W-Y. Rate of memory decline in AD is related to
education and occupation: cognitive reserve? Neurology. 1999; 53:1942–1947.
63. Beard CM, Kokmen E, Offord KP, Kurkland LT. Lack of association between
Alzheimer’s disease and education, ocupation, marital status or living arrange-
ment. Neurology. 1992; 42:2063–2069.
64. Cobb JL, Wolf PA, Au R, White R, D’Agostino RB. The effect of education on
the incidence of dementia and Alzheimer’s disease in the Framingham Study.
Neurology. 1995; 45:1707–1712.
65. Paykel ES, Brayne C, Huppert FA, Gill C, Barkley C, Gehlhaar E, Beardsall L,
Girling DM, Pollitt P, O’Connor D. Incidence of dementia in a population older
than 75 years in the United Kingdom. Arch Gen Psychiat. 1994; 51:325–332.
66. Farmer ME, Kittner SJ, Rae DS, Bartko JJ, Regier DA. Education and change
in cognitive function:The epidemiologic catchment area study. Ann Epidemiol.
1995; 5:1–7.
67. Evans DA, Beckett LA, Albert MS, Herbert LE, Scherr PA, Funkenstein HH,
Taylor JO. Level of education and change in cognitive function in a community
population of older persons. Ann Epidemiol. 1993; 3:71–77.
68. Ebly EM, Hogan DB, Parhad IM. Cognitive impairment in the nondemented eld-
erly; results from the Canadian Study of Health and Aging. Arch Neurol. 1995;
52:612–619.
69. Di Carlo A, Baldereschi M, Amaducci L, Maggi S, Grigoletto F, Scarlato G,
Inzitari D. Cognitive impairment without dementia in older people: prevalence,
vascular risk factors, impact on disability. The Italian Longitudinal Study on
Aging. J Am Geriatr Soc. 2000; 48:775–782.
70. Stern Y, Alexander GE, Prohovnik I, Mayeux R. Inverse relationship between
education and parietotemporal perfusion deficit in Alzheimer’s disease. Ann
Neurol. 1992; 32:371–375.
71. Stern Y, Alexander GE, Stricks L, Link B, Mayeux R. Relationship between life-
time occupation and parietal flow: implications for a reserve against Alzheimer’s
disease pathology. Neurology. 1995; 45:55–60.
THE BRAIN RESERVE HYPOTHESIS 239

72. Jacobs B. Schall M, Scheibel AB. A quantitiative dendritic analysis of Wernicke’s

area in humans.II Gender, hemispheric, and environmental factors. J Comp Neu-
rol. 1993; 327:97–111.
73. Diamond MC. Enriching hereditary: the impact of the environment on the anat-
omy of the brain. New York: Free Press; London: Collier MacMillan, 1988.
74. Del Ser T, Hachinski V, Merskey H, Munoz DG. An autopsy-verified study of the
effect of education on degenerative dementia. Brain. 1999; 122:2309–2319.
75. Shea S, Stein AD, Basch CE, Langtigua R, Maylahn C, Strogatz DS, Novick L.
Independent associations of educational attainment and ethnicity with behavioral
risk factors for cardiovascular disease. Am J Epidemiol. 1991; 134:567–582.
76. Lindenstrom E, Boysen G, Nyboe J. Risk factors for stroke in Copenhagen,
Denmark.I. Basic demographic and social factors. Neuroepidemiology. 1993;
12:37–42.
77. Katzman R. Education and the prevalence of dementia and Alzheimer’s disease.
Neurology. 1993; 43:13–20.
78. Chastain RL, Joe GW. Multidimensional relations between intellectual abilities
and demographic variables. J Educ Psychol. 1987; 79:323–325.
79. Mortimer JA, Graves A. Education and other socioeconomic determinants of
dementia and Alzheimer’s disease. Neurology. 1993; 43 (Suppl. 4):39–44.
80. Grafman J, Salazar A, Weingartner H, Vance S, Amin D. The relationship of brain
tissue loss volume and lesion location to cognitive deficit. J Neurosci. 1986; 6:
301–307.
81. Plassman BL, Welch KA, Helms M, Brandt J, Page WF, Breitner JCS. Intelligence
and education as predictors of cognitive function in late life: a 50-year follow-up.
Neurology. 1995; 45:1446–1450.
82. Mayeux R. Evil forces and vulnerable brains. Neurology. 2000; 55:1428–1429.
83. Jenkins R, Fox NC, Rossor AM, Harvey RJ, Rossor MN. Intracranial volume
and Alzheimer’s disease. Evidence against the cerebral reserve hypothesis. Arch
Neurol. 2000; 57:220–224.
84. Andreasen NC, Flaum M, Swayze V 2nd, O’Leary DS, Aliger R, Cohen G,
Ehrhardt J, Yuh WT. Intelligence and brain structure in normal individuals. Am
J Psychiat. 1993; 150:130–134.
85. Wickett JC, Vernon PA, Lee DH. In vivo brain size, head perimeter, and intelligence
in a sample of healthy adult females. Pers Indiv Differ. 1994; 16:831–838.
86. Reichenbach A. Glia:Neuron Index: Review and hypothesis to account for differ-
ent values in various mammals. Glia. 1989a; 2:71–77.
87. Segal M. Dentritic spines for neuroprotection: a hypothesis. Trends Neurosci.
1995; 18:468–71.
88. Arbuckle TY, Gold D, Andres D. Cognitive functioning of older people in relation
to social and personality variables. J Psychol Aging. 1986; 1:55–62.
89. Kondo K, Niino M, Shido K. A case-control study of Alzheimer’s disease in Japan
– significance of life-styles. Dementia. 1994; 5:314–326.
90. Broe GA, Henderson AS, Creasey H, McCusker E, Korten AE, Jorm AF, Longley
W, Anthony JC. A case-control study of Alzheimer’s disease in Australia. Neurol-
ogy 1990; 40:1698–1707.
91. Fabrigoule C, Letenneur L, Dartigues JF, Zarrouk M, Commenges D, Barberger-
Gateau P. Social and leisure activities and risk of dementia: a prospective longi-
tudinal study. J Am Geriatr Soc. 1995; 43:485–490.
92. Gold DP, Andres D, Etezadi J, Arbuckle T, Schwartzman A, Chaikelson J. Struc-
tural equation model of intellectual change and continuity and predictors of
intelligence in older men. Psychol Aging. 1995; 10:294–303.
93. Laurin D, Verreault R, Lindsay J, MacPherson K, Rockwood K. Physical activity
and risk of cognitive impairement and dementia in elderly persons. Arch Neurol.
2001; 58:498–504.
240 THE AGEING BRAIN

94. Friedland RP. Epidemiology, education, and the ecology of Alzheimer’s disease.
Neurology. 1993; 43:246–249.
95. Friedland RP. Epidemiology and neurobiology of the multiple determinants of
Alzheimer’s disease. Neurobiol Aging. 1994; 15:239–241.
96. Gould E, Reeves AJ, Graziano MSA, Gross CG. Neurogenesis in the neocortex
of adult primates. Science. 1999; 286:548–552.
97. Gould E, Reeves AJ, Fallah M, Tanapat P, Gross CG. Hippocampal neurogenesis
in adult Old World primates. Proc Natl Acad Sci USA. 1999; 96:5263–5267.
98. Kempermann G, Kuhn HG, Gage FH. Experience-induced neurogenesis in the
senescent dentate gyrus. J Neurosci. 1998; 18:3206–3212.
99. Van Praag H, Kempermann G, Gage FH. Running increases cell proliferation
and neurogenesis in the adult mouse dentate gyrus. Nat Neurosci. 1999; 2:
266–270.
100. Lynn R. A nutrition theory of the secular increases in intelligence;positive
correlations between height, head size and IQ. Br J Educ Psychol. 1989; 59:
372–377.
101. Bartley AJ, Jones DW, Weinberger. Genetic variability of human brain size and
cortical gyral patterns. Brain. 1997; 120:257–269.
102. Plomin R, Neiderhiser JM. Quantitative genetics, molecular genetics, and intel-
ligence. Intelligence. 1991; 15:369–387.
 SECTION IV

CLINICAL INTERFACE
Chapter 14

WILL WE ALL DEMENT

IF WE LIVE LONG
ENOUGH?
Carol Brayne

Introduction

Would we all succumb to dementia if we lived long enough? This is a ques-

tion posed frequently by the general public to the scientific community. It
is of importance both to society, as the proportion of the population made
up by older people grows, and to individuals who fear the consequences of
prolonged lives with associated frailty. The answer, if it is answerable, will
depend on a multitude of factors. This chapter cannot review even a small
fraction of the diverse literature relevant to this question, including evidence
from biomedical research on ageing, clinical and biomedical research on
dementias and also the relevant literature on chronic diseases and causes of
death in earlier life. Instead the chapter looks at the types of evidence that
have been used to address the question and explores the question itself.

Types of Evidence

There are many different types of scientific evidence upon which to draw to
attempt to answer the question. The underlying question is whether a distinc-
tion can be made between changes observed in processes considered to be
“normal” for ageing and those that are associated with dementia.

Clinical and neuropathological studies

There are many papers in which the claim is made that data are provided sup-
porting a clear differentiation of dementia, usually of the Alzheimer’s type,
244 THE AGEING BRAIN

Figure 1. How studies which are population-based can be biased.

from normal or usual ageing. The most common of these is a cross-sectional

comparison of clinical measures or investigations, or neuropathological meas-
ures where cases of disease are compared with controls. The data often show
a very different pattern for the selected controls when compared with the
demented individuals. However, this does not take account of the selection
processes, which are different for cases and controls. This kind of study does
not really provide evidence either for or against the question because of this
unknown selection bias. When similar measurements on unselected popula-
tion-based samples are made there is usually a considerable degree of overlap
of results, even when the distributions are significantly different.1
Extreme old age provides a different type of evidence. Individuals at the
extreme of old age can be compared with either younger controls or demented
individuals. Given the lack of understanding about the determinants of sur-
vival in these atypical outliers of the population, their characteristics cannot
be extrapolated to the general population. Lack of association between neu-
ropathological lesions and ageing in a population-based clinico-pathological
study has been taken as evidence that dementia is not inevitable with age, but
such a conclusion cannot be drawn when there is selective sampling for post
mortem based upon age and dementia status.2 Studies on an unselected series
of post mortems have shown steady increases in the presence of Alzheimer
type pathology in the brain with age as well as other changes such as reduc-
tion in frontal and temporal lobe cortical thickness, lower brain weight and
change in the pattern and size of neurons and supporting cells.3,4

Prevalence and incidence studies

Combined analysis of population-based studies has been conducted to
investigate the clinical expression of dementia in extreme old age. The dif-
ficulties with standardization of criteria across age groups are discussed
WILL WE ALL DEMENT? 245

Figure. 2. The raltionship between prevalence and incidence in old-age dementia.

below. Notwithstanding these problems, such reviews have suggested that

the prevalence estimates of dementia do not rise inexorably with age in
the late ninth and tenth decades, but show a levelling off. 5 This has not
been demonstrated for incidence6 and the force of mortality at these ages
is so strong that it has been shown, using deterministic modelling, that the
incidence must continue to rise in the oldest age groups in order to create
the prevalence estimates seen.7,8

Volunteer and selected population studies

Evidence that volunteer cohorts show relatively little cognitive decline is cited
against the inevitability of dementia with age.9 These tend to be highly selec-
tive and exclude the majority of the population. In Starr’s Edinburgh study9
603 out of 10,000 individuals aged 70 and over whose notes in general prac-
tice were scrutinized, had no health problems and no medication. Of these
only 195 fulfilled the same criteria a median of 4.2 years later, i.e. under 2%.
These individuals were found to have very limited cognitive decline on clinical
measures such as Mini Mental State Examination (MMSE). However changes
have been demonstrated with more sensitive tests even after accounting for
methodological issues such as practice effects on repeated testing. 10 The
expression of dementia at extremes of age has not been adequately covered
by these cohort studies and so we do not know whether these individuals
are at considerably less risk of dementing, or whether it is merely delayed in
onset. Such studies are of great interest in tracking cognitive decline in able
groups, but cannot inform public policy issues related to the majority of the
population.
246 THE AGEING BRAIN

Longitudinal studies of representative population samples

Many studies of population-based older people have reported cognitive
decline and dementia incidence. Most of these studies have concentrated
on risk factors for incident dementia, identifying a range of risks mainly
associated with vascular pathology, which confer significantly increased risk
although not extreme relative risks. Fewer have examined actual cognitive
decline and have almost universally shown that the majority of ordinary
populations show decline in measures of cognition over time, with these
being greatest in the oldest age groups.11,12 Some variation in patterns of
decline according to educational achievement and IQ has been reported,13 but
whether education truly protects from the clinical and dementing processes
remains controversial.14 This is associated, in studies with neuropathologi-
cal follow-up, with increasing expression of a variety of features mentioned
above: atrophy, loss of synapses, vascular lesions, and lesions associated with
Lewy Body Dementia and Alzheimer’s Disease (AD).15–20

Animal models
Some natural animal models are held to be helpful in throwing light on this
question, but animals do not reach equivalent extreme old age nor suffer the
same co-morbidity as humans. Some species do develop aspects of Alzheimer
type pathology. Chimpanzees develop plaques.21 Dogs develop plaques and
tangles, and show some of the vascular changes seen in humans.22–24 But
most animals do not show the same changes. Most Alzheimer research in
animals has been conducted in manipulated animals, either genetically or with
specific lesions, and does not begin to tackle the complexity of dementia in
the aged. Some combined animal models are attempting to reproduce more
realistic models.25

The Question

The denominator and the desire for absolute answers (would we all dement
if we lived long enough?)
When people ask about the inevitability of the dementia process, the ques-
tion can arise from a desire to know about personal risk if they live to
some assumed maximum life span. This assumed maximum lifespan has
not remained constant over time26 and with confirmation of survival well
beyond a hundred years has made more systematic investigation of the physi-
cal, mental and psychological state of extreme age possible.27 However, in
most cultures the maximum life span experienced would be lower and median
life span much lower. The population to which the question is applied influ-
ences the answer.
The “all” in the question above implies an assumption of certainty. Such
terms are shunned by scientists but beloved by journalists. At a personal level,
people prefer not to have to deal with uncertainty and there is acknowledged
WILL WE ALL DEMENT? 247

Figure 3. Life expectancy. Reproduced with permission from the World Bank.

difficulty in understanding risk in the population. Presenting absolute risk

for any individual requires very robust data relevant to the populations and
cohorts studied.28 In dementia, there is a substantial literature on risk, which
at some levels looks strikingly like the literature on any chronic disease and
ageing itself, with examination of processes including inflammation, apop-
tosis, immune dysfunction, chronic infection, toxic exposures, DNA and
protein damage and antioxidant effects. These diverse mechanisms, which
are about pathophysiological processes rather than causative risk, have led
some to suggest that ageing itself does not exist — “there is no such thing as
ageing”.29
At the population level we can examine the proportion of individuals who
suffer from particular pathologies and disorders at different ages, but with-
out longitudinal information on specific individuals we cannot assume with
absolute certainty that one age group will have the same patterns as another
generation at the same age. We know that disability, although not specifi-
cally cognitive disability, appears to show cohort effects in studies from the
US.30 Two major influences on expression of a disorder at a given age are
the risk patterns for that disorder in the ages leading up to the given age, and
the forces of mortality in that community. This brings us back to the “we”
of the question. At every age, even within a population and certainly across
populations, there are different influences on who has survived to that age.
In most populations there are few very old men. The demographic profile
of the British population for men has been influenced over the last hundred
years by the First World War slaughter of young men. This has consequences
for the health profile of the survivors. Those that died were the young and
more educated, leaving behind a population of survivors with different over-
248 THE AGEING BRAIN

all characteristics from those from the generations on either side of them.31
In developing countries survival patterns and consequent genetic patterns are
heavily influenced by morbidity at young ages, and now HIV infections. In
these societies, given that average life span does not yet reach the ages where
dementia is common, and individuals with extreme life spans are exceedingly
rare, the chances of dementing are not zero, but low, and those who survive
into old age are highly selected. Even in a condition where the genetic predis-
position to dementia is known, such as Huntington’s Chorea in which inherit-
ance is dominant, there is variation in the nature and timing of onset leading
to uncertainty in predicting age of onset. If life expectancy were as short as it
was in early periods of human development, only a minority of the individuals
carrying the Huntington’s Chorea mutations would have developed dementia,
and it therefore would not have been inevitable.
Specific interrelationships might be seen in genetic profiles in developed
countries. An example of this is the reported relationship between butyryl-
cholinesterase and apolipoprotein E alleles as a risk for AD in men in a
meta-analysis.32 This pattern would not necessarily be replicable in other
populations.
Sex differences in the expression of dementia can also be influenced by
survival patterns.33 Men tend to live shorter lives than women, even without
the world war experiences noted above, and older women are known to suffer
from more disability at a given age than older men in most populations where
this has been examined.19 Thus, at present, men in the oldest age groups are
healthier than surviving women and may well be at lower risk from dementia.
To be recognized as demented, particularly with AD, we have to survive
through earlier cognitive decline to fulfil diagnostic criteria. But those people
who dement die at a faster rate than those who are not demented, even within
the same age group.34 Cognitive decline is also associated with increased mor-
tality.19 Thus at all ages, and particularly in the oldest age groups (where the
general force of mortality is very great), the survivors from any given age are
less likely to be those who are demented and those who are soon to become
demented.
The cultures where dementia has, through painstaking research using
clinically validated comparative methods,35 been shown to be less common
are India and Nigeria when compared with population contemporaries in
the United States.36,37 In these cultures the pattern and genetic risk for AD
appear to be different, and suggest either survival effects or different gene-
environment risk interaction or possibly both. A study of a locality in India
revealed the much lower proportions of individuals with the known risk gene
apolipoprotein ε4 than in the western populations in whom the risk was
originally identified. However the risk associated with this allele was similar
to that observed in the west.38 In Nigeria the risk allele was more common,
but there was no associated risk of dementia.36 Attenuation with age of the
effects of the known risk factor for AD apolipoprotein ε4 supports the need
for examination of the impact of risk at different ages.39,40
WILL WE ALL DEMENT? 249

Even when longitudinal studies are done, and incidence is found to be

lower, the possibility remains that the majority of the population has not
reached the age of maximum risk for late onset dementing conditions as their
life expectancies are not sufficiently long.7,8 In those countries where life
expectancy has increased dramatically over the past decades, such as Japan,
the prevalence of dementia is remarkably similar to Western societies. The
relative proportions of different clinically diagnosed subtypes of dementia
may vary as might be predicted given different levels of vascular risk across
populations.41

The disorder itself — what is dementia?

Implicit in the “would we all dement?” is the differentiation of usual from
successful and unsuccessful. What is usual with age and what unusual, what
normal and what abnormal? Basic epidemiological and statistical textbooks
on the distributions deal well with the semantic issues raised here. Last42
describes “normal” as having three distinct meanings and points out that
conceptual difficulties may arise if these different meanings are not specified
or if the area of overlap is not clearly understood (see Box).

1. Within the usual range of variation in a given population or popula-

tion group, or frequently occurring in a given population or group.
In this sense, “normal” is frequently defined as, “within a range
extending from two standard deviations below the mean to two
standard deviations above the mean”, or “between specified (e.g.,
the 10th and 90th percentiles of the distribution).

2. In good health, indicative or predictive of good health, or conducive

to good health. For a diagnostic or screening test, a “normal” result
in one in a range within which the probability of a specific disease
is low.

3. Gaussian distribution [i.e. bell curve].

From Last .42

These are rarely specified in discussions about dementia, and even less in
discussions about the cognitive impairment seen in many older people that
can precede dementia. The dementia process has been described in western lit-
erature over the ages, at its most basic as a loss of mind and more recently as
‘a chronic or persistent disorder of the mental processes marked by memory
disorders, personality changes, impaired reasoning etc., due to brain disease
or injury’ (Concise Oxford Dictionary of Current English, 1990). Frailty
associated with ageing is acknowledged in many cultures, but dementia is
250 THE AGEING BRAIN

not necessarily recognised. The definition of dementia has varied with time
even within western societies, but has now been operationalised within the
International Classification of Diseases of the World Health Organisation43
and the Diagnostic and Statistical Manual of the American Psychiatric Asso-
ciation.44
To make a diagnosis of the syndrome of dementia during life requires
information from several different axes, the most important are given in the
dictionary definition and include cognition, function, behaviour and mood.
To recognize abnormal function requires personal, community and societal
expectations and norms. These axes themselves are not independent and one
will predict decline on another. For example functional ability as measured
by activities of daily living can predict the onset of dementia45 despite the
fact that functional decline is supposed to be a consequence of the cognitive
impairment of dementia. Once the diagnosis of dementia is made, there is
usually an attempt to make a specific subtype diagnosis. To do this requires
a further combination of history, examination and investigation to identify
a range of possible underlying pathologies, of which there are many46 with
vascular47 and Alzheimer’s being dominant. Its onset, type of progression,
comorbid conditions and particular findings on investigation, which might
eventually include the brain itself, can then characterise the dementia. The
meaning of the word and the diagnosis “dementia” can thus mean very dif-
ferent things according to whether it has been based on a single measure, such
as cognition, or a combination such as cognition and function, or the whole
range including detailed investigations and exclusions.
It is clear that applying different methods identifies different individu-
als,48,49 and that subtype diagnosis in ordinary settings is not necessarily good
at identifying the true underlying pathologies.50 It is also clear that diagnostic
criteria are varied according to the age at which they are applied.51,52 Patterns
considered to be abnormal in younger age groups are allowable in older age
groups. The assessment of centenarians for dementia would be less stringent
than that for people in their sixties. Relaxation of criteria with age is illus-
trated by the methods used for the French study of extreme old age where to
be assumed non-demented an individual had to have a reported meaningful
verbal exchange recently — such as “that’s a nice dress you are wearing”.27
Illustration of this principle is illustrated by the NINCDS ADRDA neu-
ropathological criteria for AD in which the absolute pathologies shown were
given different cutpoints, with more Alzheimer type pathology allowable in
the older age groups.53
The Consortium to Establish a Register for Alzheimer’s Disease group
(CERAD) produced consensus criteria for definite Alzheimer’s disease which
require a clinical diagnosis of dementia and a sufficient level of Alzheimer
type pathology to warrant a diagnosis of AD, based on the demonstration of
plaques.54 This demonstrates the dissociation of clinical evidence from bio-
logical evidence in that one could be discounted in the presence or absence of
the other. Thus, if the question “Will we all dement if we lived long enough?”
WILL WE ALL DEMENT? 251

was rephrased to “Will we all demonstrate neuropathological changes in our

brains associated with dementia if we live long enough?” the answer would be
a categorical yes. All brains over a certain age will show some atrophy, prob-
ably some vascular lesions and some Alzheimer type pathology,2,3 with some
studies reporting on the added risk of experiencing dementia during life with
a combination of pathologies,55,56 and others reporting on the close similarity
of the Alzheimer type pathology observed in the very old non-demented when
compared with established cases of AD.57
The criteria for dementia and its subtypes are dominated by western clini-
cal thinking and methods of assessment, and these methods and assessment
have been adopted and adapted by many different cultures as described
above, within the model of westernised medical training and cultural expec-
tations. Where the methods have been carefully adapted and cross-validated
across cultures, systematic differences have been demonstrated in the propor-
tion of people diagnosed at different ages and developing dementia.36,37 The
question of whether this is related to survival to the ages of risk, and mortal-
ity differentials for those with incipient dementia is not yet fully answered.
It does not appear to be due to lack of risk for the basic neuropathological
changes, since, despite earlier reports to the contrary, we now know that there
are Alzheimer type pathological changes in the brains of older Africans.58
Despite the vast evidence collected and published to date, there is still
uncertainty about which brain pathology or pathologies underlie clinical
dementia seen during life.59–61 The potential mechanisms such as inflamma-
tion are mentioned above, but neuropathological criteria for the different
dementias include a variety of specific changes. These range from neuronal
loss, Alzheimer type pathologies to white matter pallor, sclerosis62 vascular
abnormalities from accumulation of proteins in the cell walls63 to thrombosis
and ischaemic damage. The issue of what is normal and abnormal ageing is
important because most of these changes are found with increasing frequency
in ageing brains. Some have suggested integrated hypotheses with many trig-
gering factors leading to common final pathways.64 How these processes
might be staged remains uncertain. It has been suggested that pure neuronal
loss can be associated with dementia and thus in the absence of all prevent-
able pathologies, this would be the likely dementia of extreme old age.65
Whether individuals exhibiting some, or all, of these pathologies but
without having expressed dementia are at an early and unrecognised stage of
dementia remains a question. Many think they are66,67 and some suggest that
the in vivo clinical measures have been too insensitive to recognise abnor-
mality. However we have shown above that sensitive measures demonstrate
change in most older people. Thus the argument becomes circular as this
would bring much greater proportions of the population into the abnormal
range. There is substantial cognitive impairment which does not reach diag-
nostic criteria for dementia, and this increases with age even after accounting
for co-morbidity and sensorimotor problems.11,68,69 Mild cognitive impair-
ment has been developed into a medicalised category, with its own criteria.
252 THE AGEING BRAIN

This appears to be preparing the ground for trials of interventions which

could be aimed at substantial proportions of the population, potentially from
young ages, and could be seen as part of a more general trend of medicalisa-
tion of what used to be considered normal processes.70

What is “old”?
There is continuing, unresolved, debate about the nature of ageing itself
and what might be the best measure of biological age, if indeed such a thing
exists.29 Individuals who survive past a century have been described as having
been youthful in their young old age.27 This suggests that, at least for some
aged individuals, there are mechanisms at play that are protecting them from
the usual ageing experienced by the vast majority of older people. Older peo-
ple themselves have diverse views about the reasons for their successful old
age in our own local population studies, these tend to emphasise moderation
in lifestyle, with continuing engagement in social and physical activities. As
mentioned above, there are other aspects of ageing, beyond the biological,
that might be considered such as societal, chronological and psychological.
Enforced retirement of certain careers is based on the assumption that certain
powers diminish with age in a way that could jeopardize the particular job. If
a society expects very high functioning from all individuals, those with cogni-
tive decline who started low will arrive at the dementia threshold earlier than
others, and at certain ages a very high proportion will reach fixed criteria.
An illustration of this is the fact that insurance companies tend to use scale
measures to define functional and cognitive disability. If a threshold score of
23/24 on the Mini Mental State Examination (MMSE) were to be used, the
majority of women aged 85 and over would be declared to have dementia in
England and Wales.69

Longevity and dementia

How long would be long enough? This has been discussed above in relation
to variation in life expectancy in the world. There is a hint of consideration of
quality in the term “enough”. Enough may be too much. The question here
might be an implicit challenge to the medical model of extension of life at all
cost, on the basis that quality of life is not necessarily associated with quantity.
This is a truism, but in practical application of western medicine many indi-
viduals have their lives extended without quality,71 and possibly with greater
risk of acquiring a dementing condition. It has been said “seeking the biologi-
cal basis of ageing is to continue to pursue technological solutions to age old
problems, without addressing the fundamental question of how to improve the
experience of the end of life”.72 Consequently the likelihood of extension of
life has made the chance of impairing quality of life for some people greater.

What would we want for ourselves?

Those individuals diagnosed at the early stage of the dementia process, along
with their carers, seek support and interventions that at best reverse or, at
WILL WE ALL DEMENT? 253

least, slow its progression and manifestations. Those diagnosed at later stages
may not wish for reversal of progression unless it were guaranteed to proceed
to a sufficient level to improve quality of life. It is possible to argue that there
are situations in which informed choice for carers and sufferers might include
a rejection of some types of treatment. There might not be agreement between
the carers and patients.73 Examples might be if the improvement were to be
associated with greater insight into impairments and longer life expectancy,
or to improve specific cognitive impairment without improving behavioural
difficulties. Relatively little discussion and research has been carried out on
public attitudes to treatment of dementia and effects of treatment, partly
because trials only assess relatively limited timescales, specific types of inter-
vention and limited aspects of treatment effect. The concentration on identify-
ing single types of dementia, with single causes and simple remedies, mostly in
the form of medication74 has led to an expectation of cure. The media support
this emphasis which, while entirely understandable, may not provide enough
information on which to base informed policies for prevention and care in
whole populations. In view of the increasing debate about euthanasia, the
right to refuse intervention and active rejection of maintenance of life without
quality are likely to be a focus for future debate.

Conclusion

This chapter has explored the assumptions implicit in the debate about the
relationship of ageing to dementia. Examination of the strength of the evi-
dence and the assumptions inherent in the question are of relevance to policy
makers, clinicians and researchers. However, these issues are possibly of
greater significance to societies than the original question itself: the impact
of changing patterns of ageing with more or less fit individuals reaching old
age; the high likelihood of having brain pathologies at extreme old age along
with the limitations of a binary approach (you either have it or you don’t)
75,76; the observation that the majority of older people do experience some

cognitive decline with age paralleled by increasing changes in the brain; and
the overall impact, including health, social and economic, on society of any
preventive treatments which may be generated.

References

1. Brayne C, Calloway P. Is Alzheimer’s disease distinct from normal ageing? Lan-

cet. 1988; 1:1265–1267.
2. MRC CFAS Neuropathology Group: Ince P, Matthews F, Brayne C, Esiri M.
Pathological correlates of late-onset dementia in a multi-centre, community-based
population in England and Wales. Lancet. 2001; 357:169–175.
3. Miller FD, Hicks SP, D’Amato CH, Landis JR. A descriptive study of neuritic
plaques and neurofibrillary tangles in an autopsy population. Am J Epidemiol.
1984; 3:331–341.
254 THE AGEING BRAIN

4. Terry RD, De Teresa R, Hansen LA. Neocortical cell counts in normal human
adult aging. Ann Neurol. 1987; 21:530–539.
5. Ritchie K, Kildea D. Is senile dementia “age related” or “ageing related”? Evi-
dence from meta-analysis of dementia prevalence in the oldest old. Lancet. 1995;
346:931–934.
6. Jorm AF, Jolley D. The incidence of dementia: a meta-analysis. Neurology. 1998;
51:728–733.
7. McGee MA, Brayne C. The impact on prevalence of dementia in the oldest age
groups of differential mortality patterns: a deterministic approach. Int J Epide-
miol. 1998; 27:87–90.
8. McGee MA, Brayne C. Exploring the impact of prevalence and mortality on inci-
dence of dementia in the oldest old: the sensitivity of a deterministic approach.
Neuroepidemiology. 2001; 20:221–224.
9. Starr JM, Deary IJ, Inch S, Cross S, MacLennan WJ. Age-associated cognitive
decline in healthy old people. Age Ageing. 1997; 26:295–300.
10. Rabbitt P, Lowe C. Patterns of cognitive ageing. Psychol Res. 2000; 63: 308–
316.
11. Brayne C, Spiegelhalter DJ, Dufouil C, Chi LY, Dening TR, Paykel ES, O’Connor
DW, Ahmed A, McGee MA, Huppert FA. Estimating the true extent of cognitive
decline in the old old. J Am Geriatr Soc. 1999; 47:1283–1288.
12. Cullum S, Huppert FA, McGee MA, Denning T, Ahmed A, Paykel ES, Brayne C.
Decline across different domains of cognitive function in normal ageing: results
of a longitudinal population based study using CAMCOG. Int J Ger Psychiat.
2000; 15:853–862.
13. Leibovici D, Ritchie K, Ledesert B, Touchon J. Does education level determine
the course of cognitive decline? Age Ageing. 1996; 25:392–397.
14. Christensen H, Hofer SM, Mackinnon AJ, Korten AE, Jorm AF, Henderson AS.
Age is not kinder to the better educated: absence of an association investigated
using latent growth techniques in a community sample. Psychol Med. 2001; 31:
15–28.
15. Masliah E, Mallory M, Hansen L, De Teresa R, Terry RD. Quantitative synaptic
alterations in the human neocortex during normal aging. Neurology. 1993; 43:
192–197.
16. Hansen LA, Terry RD. Plaque-only Alzheimer disease is usually the lewy body
variant and vice versa. J Neuropath Exp Neurol. 1993; 52:648–654.
17. Xuereb JH, Brayne C, Dufouil C. Neuropathological findings in the very old.
Results from the first 101 brains of a population-based longitudinal study of
dementing disorders. Ann NY Acad Sci. 2000; 47:490–496.
18. Green MS, Kaye JA, Ball MJ. The Oregon brain ageing study: neuropathology
accompanying healthy ageing in the oldest old. Neurology. 2000; 54:105–113.
19. MRC CFAS: Neale R, Brayne C, Johnson A. Cognition and survival: an explora-
tion in a large multicentre study of the population aged 65 years and over. Int J
Epidemiol. 2001; 30:1383–1388.
20. Kalaria RN, Ballard CG, Ince PG, Kenny RA, McKeith IG, Morris CM, Brien JT,
Parry EK, Perry RH, Edwardson JA. Multiple substrates of late-onset dementia:
implication for brain protection. Novart Fdn Symp. 2001; 235:49–60.
21. Gearing M, Rebeck GW, Hyman BT, Tigges J, Mirra SS. Neuropathology and
apolipoprotein E profile of aged chimpanzees: implications for Alzheimer disease.
Proc Natl Acad Sci USA. 1994; 91:9382–9386.
22. Cummings BJ, Head E, Ruehl W, Milgram NW, Cotman CW. The canine as
an animal model of human aging and dementia. Neurobiol Aging. 1996; 17:
259–268.
23. Papaioannou N, Tooten PC, van Ederen AM, Bohl JR, Rofina J, Tsangaris T,
Gruys E. Immunohistochemical investigation of the brain of aged dogs. I. Detec-
WILL WE ALL DEMENT? 255

tion of neurofibrillary tangles and of 4-hydroxynonenal protein, an oxidative

damage product, in senile plaques. Amyloid. 2001; 8:11–21.
24. Head E, Torp R. Insights into Abeta and presenilin from a canine model of human
brain aging. Neurobiol Dis. 2002; 9:1–10.
25. Trojanowski JQ. Neuropathological verisimilitude in animal mmodels of Alzhei-
mer’s disease. Am J Pathol. 2002; 160:409–411.
26. Kirkwood TBL. Is there a biological limit to the human life span? In: Robine,
JM, Vaupel JW, Jeune B, Allard M, editors. Longevity: to the limits and beyond.
Berlin: Springer Verlag, 1997; 69–76.
27. Allard M, Robine JM. Les centenaries francais. Paris: Serdie, 2000.
28. Calman K. Cancer: science and society and the communication of risk. Br Med J.
1996; 313:799–802.
29. Peto R, Doll R. There is no such thing as aging. Br Med J. 1997; 315:1030-1032.
30. Reynolds SL, Crimmins EM, Saito Y. Cohort differences in disability and disease
presence. Gerontologist. 1999; 38:578–590.
31. Whalley LJ, Deary IJ. Longitudinal cohort study of childhood IQ and survival up
to age 76. Br Med J. 2001; 322: 819.
32. Lehmann DH, Williams J, McBroom J, Smith AD. Using meta-analysis to explain
the diversity of results in genetic studies of late-onset Alzheimer’s disease and to
identify high-risk subgroups. Neuroscience. 2001; 108:541–554.
33. Hill GB, Forbes WF, Lindsay J. Life expectancy and dementia in Canada: the
Canadian study of health and aging. Chronic Dis Can. 1997; 18:166–167.
34. Dewey ME, Saz P. Dementia, cognitive impairment and mortality in persons aged
65 and over living in the community: a systematic review of the literature. Int J
Geriatr Psych. 2001; 16:751–761.
35. Fillenbaum GG, Chandra V, Ganguli M, Pandav R, Gilby JE, Seaberg EC, Belle
S, Baker C, Echement DA, Nath LM. Development of an activities of daily living
scale to screen for dementia in an illiterate rural older population in India. Age
Ageing. 1999; 28:161–168.
36. Hendrie HC, Ogunniyi A, Hall KS, Baiyewu O, Unverzagt FW, Gureye O, Gao
S, Evans RM, Ogunseyinde AO, Adeyinka AO, Musick B, Hui SL. Incidence of
dementia and Alzheimer disease in 2 communities: Yoruba residing in Ibadan,
Nigeria and African Americans residing in Indianapolis, Indiana. J Amer Med
Assoc. 2001; 14:739–747.
37. Chandra V, Pandav R, Dodge HH, Johnston JM, Belle SH, DeKosky ST, Ganguli
M. Incidence of Alzheimer’s disease in a rural community in India: the Indo-US
study. Neurology. 2001; 57:985–989.
38. Ganguli M, Chandra V, Kamboh MI, Johnston JM, Dodge HH, Thelma BK,
Juyal RC, Pandav R, Belle SH, DeKosky ST. Apolipoprotein E polymorphism and
Alzheimer disease: The Indo US Cross-National Dementia Study. Arch Neurol.
2000; 57:824–830.
39. Sulkava R, Kainulainen K, Verkkoniemi L, Ninisto L, Sobel E, Davanipour Z,
Polvikoski T, Haltia M, Kontula K. APOE alleles in Alzheimer’s disease and vas-
cular dementia in a population aged 85+. Neurobiol Aging. 1996; 17:373–376.
40. Rubinsztein DC, Easton DF. Apolipoprotein E genetic variation and Alzheimer’s
disease. A meta-analysis. Dement Geriatr Cogn Disord. 1999; 10:199–209.
41. Ikeda M, Hokoishi K, Maki N, Nebu A, Tachibana N, Komori K, Shigenobu K,
Fukuhara R, Tanabe H. Increased prevalence of vascular dementia in Japan: a
community-based epidemiological study. Neurology. 2001; 57:839–844.
42. Last JM. A dictionary of epidemiology. 4th edition. Oxford, New York: Oxford
University Press, 2001.
43. Worls Health Organization. International statistical classification of diseases and
related health problems. – Tenth Revision. Geneva: Worls Health Organization,
1994.
256 THE AGEING BRAIN

44. American Psychiatric Assocation: Diagnostic and statistical manual of mental

disorders, 4th ed. Wastinghton, DC: American Psychiatric Association, 1994.
45. Barberger-Gateau P, Dartigues JF, Letenneur L. Four instrumental activities of
daily living score as a predictor of one-year incident dementia. Age Ageing. 1993;
22:457–463.
46. Ince PG, McArthur FK, Bjertness E, Torvik A, Candy JM, Edwardson JA. Neu-
ropathological diagnoses in elderly patients in Oslo: Alzheimer’s disease, Lewy
Body disease, Vascular lesions. Dementia. 1995; 6:162–168.
47. Roman GC, Tatemichi TK, Erkinjuntti T, Cummings JL, Masdeu JC, Garcia JH,
Amaducci L, Orgogozo JM, Brun A, Hofman A. Vascular dementia: diagnostic
criteria for research studies: report of the NINDS-AIREN International Work-
shop. Neurology. 1993; 43:250–260.
48. Erkinjuntti T, Ostbye T, Steenhuis R, Hachinski V. The effect of different
diagnostic criteria on the prevalence of dementia. N Engl J Med. 1997; 337:
1667–1674.
49. Thomas VS, Darvesh S, MacKnight C, Rockwood K. Estimating the prevalence
of dementia in elderly people: a comparison of the Canadian Study of Health and
Aging and National Population Health Survey approaches. Int Psychogeriatr.
2001; 13 (Suppl 1): 169–175.
50. Gilleard CJ, Kellett JM, Coles JA, Millard PH, Honavar M, Lantos PL. The St
Georges dementia bed investigation study: a comparison of clinical and pathologi-
cal diagnoses. Acta Psychiat Scand. 1992; 85:264–269.
51. Brayne C. Clinicopathological studies of the dementias from an epidemiological
viewpoint. Br J Psychiat. 1993; 162:439–446.
52. Hansen LA, Terry RD. Position paper on diagnostic criteria for Alzheimer disease.
Neurobiol Aging. 1997; 18:S71–73.
53. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical
diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group
under the auspices of Department of Health and Human Services Task Force on
Alzheimer’s Disease. Neurology. 1984; 34:939–944.
54. Mirra SS, Heyman A, McKeel D, Sumi SM, Crain BJ, Brownlee LM, Vogel FS,
Hughes JP, van Belle G, Berg L. The Consortium to Establish a Registry for
Alzheimer’s disease (CERAD). Part II. Standardization of the neuropathologic
assessment of Alzheimer’s disease. Neurology. 1991; 41:479–486.
55. Tomlinson BE, Blessed G, Roth M. Observations on the brains of demented old
people. J Neurol Sci. 1970; 11:205–242.
56. Snowdon DA, Greiner LH, Mortimer JA, Riley KP, Greiner PA, Markesbery WR.
Brain infarction and the clinical expression of Alzheimer’s Disease. The Nun
study. J Amer Med Assoc. 1997; 277:813–817.
57. Arriagada PV, Marzloff K, Hyman BT. Distribution of Alzheimer-type pathologi-
cal changes in nondemented elderly individuals matches the pattern in Alzheim-
er’s disease. Neurology. 1992;4 2:1681–1688.
58. Ogeng’o JA, Cohen DL, Sayi JG, Matuja WB, Chande HM, Kitinya JN, Kimani
JK, Freidland RP, Mori H, Kalaria RN . Cerebral amyloid beta protein deposits
and other Alzheimer lesions in non-demented elderly east Africans. Brain Pathol.
1996; 6:101–108.
59. Arriagada PV, Growdon JH, Hedley Whyte T, Hyman BT. Neurofibrillary tan-
gles but not senile plaques parallel duration and severity of Alzheimer’s disease.
Neurology. 1992; 42:631–639.
60. Bancher C, Jellinger KA. Neurofibrillary tangle predominant form of senile
dementia of Alzheimer type: a rare subtype in very old subjects. Acta Neuropath.
1994; 88:565–570.
61. Mackenzie IRA, McLachlan RS, Kubin CS, Miller LA. Prospective neuropsycho-
logical assessment of nondemented patients with biopsy proven senile plaques.
Neurology. 1996; 46:425–429.
WILL WE ALL DEMENT? 257

62. Dickson DW, Davies P, Bevona C, Van Hoeven KH, Factor SM, Grober E, Aron-
son MK, Crystal HA. Hippocampal scelosis: a common pathological feature of
dementia in the very old (> or = 80 years of age) humans. Acta Neuropath. 1994;
88:212–221.
63. Ellis RJ, Olichney JM, Thal LJ, Mirra SS, Morris JC, Beekly D, Heyman A. Cer-
ebral amyloid angiopathy in the brains of patients with Alzheimer’s disease: the
CERAD experience part XV. Neurology. 1996; 46:1592–1596.
64. Hardy JA, Mann DMA, Wesler P, Winblad B. An integrative hypothesis concern-
ing the pathogenesis and progression of Alzheimer’s disease. Neurobiol Aging.
1986; 7:489–502.
65. Terry RD, Katzman R. Lifespan and synapses: will there be a primary senile
dementia? Neurobiol Aging. 2001; 22:347–248.
66. Linn RT, Wolf PA, Bachman DL, Knoefel JE, Cobb JL, Belanger AJ, Kaplan EF,
D’Agostino RB. The preclinical phase of probable Alzheimer’s disease. A 13-year
prospective study of the Framingham cohort. Arch Neurol. 1995; 52:485–490.
67. Morris JC, Storandt M, Miller JP, McKeel DW, Price JL, Rubin EH, Berg L.
Mild cognitive impairment represents early-stage Alzheimer disease. Arch Neurol.
2001; 58:397–405.
68. Ebly EM, Hogan DB, Parhad IM. Cognitive imairment in the nondemented eld-
erly. Results from the Canadian Study of Health and Aging. Arch Neurol. 1995;
52:612–619.
69. Brayne C, Nickson J, Johnson A. Cognitive function and dementia in six areas of
England and Wales: the distribution of MMS and the prevalence of GMS organic-
ity level in the MRC CFA Study. Psychol Med. 1998; 28:319–335.
70. Moynihan R, and Smith R. (2002) Too much medicine? Br Med J. 2002; 324:
859–860.
71. Shapin S, Martyn C. How to live forever: lessons of history. Br Med J. 2000; 321:
1580–1582.
72. Tallis RC. Brains and minds: a brief history of neuromythology. J Roy Coll Phys
Lond. 2000; 34:563–567.
73. Novella JL, Jochum C, Jolly D, Morrone I, Ankri J, Bureau F, Blanchard F.
Agreement between patients’ and proxies’ reports of quality of life in Alzheimer’s
disease. Qual Life Res. 2001; 10:443–452.
74. Gallagher M, Gill M, Baxter MG, Bucci DJ. Semin Neurosci. 1994; 6:351–358.
75. Wainwright NWH, Surtees PG, Gilks WR. Diagnostic boundaries, reasoning and
depressive disorder. I Development of a Probabilistic model for public health
psychology. Psychol Med. 1997; 27:835–845.
76. 76. Surtees PG, Wainwright NWH, Gilks WR. Diagnostic boundaries, reasoning
and depressive disorder. II. Diagnostic complexity and depression: time to allow
for uncertainty. Psychol Med. 1997; 27:847–860.
Chapter 15

DETECTING
ALZHEIMER’S DISEASE AT
THE PRE-SYMPTOMATIC
STAGES
Gary W Small

Introduction

As people liver longer, the risk for developing Alzheimer’s disease (AD)
increases dramatically. In fact, the incidence appears to double every five
years after age 60 years, suggesting that if people lived long enough, they
would all develop the disease by a certain age. Although AD is the most com-
mon cause of late-life dementia, other causes, particularly vascular disease, do
contribute to the occurrence of dementia. In fact, the burden of such vascular
decline appears to contribute to a greater portion of dementia cases in the
upper age groups.
Whether it is pure AD, pure vascular dementia, or something along the
continuum, these cases of dementia progress with time. The underlying lesions
reach a threshold such that they lead to cognitive decline that interferes with
daily life. With Alzheimer’s dementia, this slow insidious decline represents
an accumulation of pathological features and declining neurotransmitter
functions that begin well before the clinician can confirm a clinical diagnosis
in practice.

Adapted in large part from: Small GW. Structural and functional imaging of Alzhe-
imer’s disease. In: Davis KL, Charney D, Coyle JT, Nemeroff C, editors. Neuropsy-
chopharmacology: the fifth generation of progress. Philadelphia: Lippincott, Williams
and Wilkins, 2002; 1231–1242.
260 THE AGEING BRAIN

Focus on Early Detection

With the realization that the neuropathological changes of AD begin to accu-

mulate perhaps decades before the disease is obvious clinically, studies have
emphasized recruitment of subjects at time points years or decades before a
physician begins to focus on early detection of AD at clinical stages. These
studies might confirm a clinical diagnosis of probable AD.1 The ultimate
goal is to develop tools to identify pre-symptomatic candidates for beginning
preventive pharmacological treatments before extensive neuronal damage
develops. Brain imaging has become an important tool for the development
of surrogate markers that will effectively identify people with only mild
cognitive losses who are likely to progress in their cognitive loss and eventu-
ally develop the full dementia syndrome of AD. As novel, disease-modify-
ing agents emerge, these surrogate brain-imaging markers will be critical in
determining drug efficacy and facilitating drug development in both animal
models and human studies.

Diagnostic Categories

Clinical investigators have developed definitions for categorical pre-symp-

tomatic stages that assist in clinical trials and communicating staging levels.
Several diagnostic entities have been described in efforts to better characterize
age-related cognitive decline. The mildest form of age-related memory decline
is known as age-associated memory impairment (AAMI),2 characterized by
self-perception of memory loss and a standardized memory test score > 1SD
below the aged norms. In people 65 years of age or older, its estimated
prevalence is 40%, afflicting approximately 16 million people in the United
States.3 Only about 1% of such cases will develop dementia each year. The
term AAMI has generated controversy since many people question the specific
criteria and whether they define a stable or declining entity.
A more severe form of memory loss is mild cognitive impairment (MCI),
often defined by significant memory deficits without functional impairments.
People with MCI show memory impairment that is > 1.5 SDs below aged
norms on such memory tasks as delayed paragraph recall.4 Approximately
10% of people 65 years or older suffer from MCI, and nearly 15% develop
AD each year.4,5 This condition also has generated controversy. Some experts
consider MCI a risk state rather than a diagnostic entity. Despite such con-
troversy, MCI appears to be a useful concept and may respond favorably to
current symptomatic treatments. Brain imaging studies of pre-symptomatic
AD focus on both these forms of age-related memory decline.
DETECTING ALZHEIMER’S DISEASE AT THE PRE-SYMPTOMATIC STAGES 261

Evidence for Pre-Symptomatic Changes

Several areas of research have contributed to the idea that pre-symptomatic

conditions do exist, including neuropathological, neuroimaging, and clini-
cal investigations. Taken together, this work supports the notion that the
dementing process leading to AD begins years before a clinical diagnosis of
probable AD can be confirmed.6 Post-mortem studies of non-demented older
people7 indicate that tangle density in healthy ageing correlates with age, but
that some cases demonstrate widely distributed neuritic and diffuse plaques
throughout neocortical and limbic structures. Other studies8 have found that
neurofibrillary tangle density increases in some individuals, presumably those
who will eventually develop AD very early in adult life, perhaps even by the
fourth decade. The diffuse amyloid deposits in middle-aged non-demented
subjects are consistent with an early or pre-symptomatic stage of AD and
suggest that the pathological process progresses gradually, taking 20 to 30
years to proceed to the clinical manifestation of dementia.9 Other supportive
evidence includes findings that linguistic ability in early life predicts cognitive
decline in late life.10 High diffuse plaque density in non-demented older per-
sons has been observed in the entorhinal cortex and inferior temporal gyrus,
in association with acetylcholinesterase fibre density.11 Evidence from animal
models also supports compromised hippocampal cholinergic transmission
during ageing.12 Studies of glucose metabolic rates using positron emission
tomography (PET)6,13,14 indicate lower regional brain metabolism in middle-
aged and older persons with a genetic risk (apolipoprotein E [ApoE ε4]) +ve
status, lending further support for a prolonged pre-symptomatic AD stage.

Structural Imaging

Computerized tomography and magnetic resonance imaging

The largest body of data comes from studies of structural imaging modalities
in that clinical investigators have had the greatest access to these technologies.
Studies of early detection logically follow from initial work demonstrating the
differential diagnostic utility of a brain-imaging marker. For structural imag-
ing, particularly magnetic resonance imaging (MRI), data have emerged on
the use of regional atrophy patterns for the positive diagnosis of AD and other
neurodegenerative disorders. Studies without neuropathological confirmation
report the utility of medial temporal lobe atrophy, particularly hippocampal
atrophy, on computerized tomography (CT) or MRI for the clinical diagnosis
of AD.15 Some but not all quantitative MRI studies indicate that white mat-
ter hyperintensities correlate with neuropsychological functioning in both
healthy elderly persons and demented patients.16,17 Other studies indicate loss
of cerebral gray matter,18 hippocampal and parahippocampal atrophy,19 and
lower left amygdala and entorhinal cortex volumes20 in patients with AD.
In differentiating AD from older normal controls, the sensitivity of various
262 THE AGEING BRAIN

medial temporal atrophy measures ranges from 77% to 92%, with specifici-
ties ranging from 49% to 95%.21–23 In older MCI patients, hippocampal
atrophy predicts subsequent conversion to AD.24 Of various analytic meth-
ods, computerized volumetric techniques are most accurate, but are currently
labor-intensive and not widely available.
A modified negative-angle axial view designed to cut parallel to the ante-
rior-posterior plane of the hippocampus has been used to assess hippocampal
volume using CT or MRI.15 Such hippocampal atrophy is a sensitive and
specific predictor of future AD in patients with MCI. Baseline hippocam-
pal ratings accurately predicted decliners with an overall accuracy of 91%.
Neuropathological studies find that the sites of maximal neuronal loss for
both AD and MCI are in the CA1, subiculum, and entorhinal cortex.15 Hip-
pocampal atrophy also has been found to predict future cognitive decline in
older persons without cognitive impairment followed for nearly four years.
Visual assessments of medial temporal lobe atrophy on coronal MRI sections
show significant correlations between estimated and stereologically measured
volumes.25 Because the latter is much more labor-intensive, visual readings
may be an alternative approach with greater efficiency.
The hippocampus and the temporal horn of the lateral ventricles also may
serve as antemortem AD markers in mildly impaired patients (mean MMSE
score of 24).26 While hippocampal atrophy may distinguish AD from normal
ageing, such atrophy may be non-specific, occurring in other dementing dis-
orders.27 Magnetic resonance imaging hippocampal atrophy measures are
not as sensitive as PET glucose metabolism measures, which begin decreasing
before memory decline onset.28 The presence of MRI white matter hyperin-
tensities does not improve diagnostic accuracy since they occur both in AD
and healthy normal elderly.29,30
The entorhinal cortex (EC), a region involved in recent memory perform-
ance, is one of the earliest areas to accumulate NFTs.8 Histological bounda-
ries of the EC from autopsy-confirmed AD patients and controls have been
used to validate a method for measurement of EC size relying on gyral and
sulcal landmarks visible on MRI.31 Such measures may be additional early
AD detection markers.
Several studies have addressed the interaction between regional atrophy
and ApoE genotype. Increasing dose of ApoE ε4 allele was associated with
smaller hippocampal, entorhinal cortical, and anterior temporal lobe vol-
umes in already demented patients.32 A study of non-demented older persons
found an association between ApoE ε4 dose and a larger left than right hip-
pocampus.33 Combining medial temporal measures with other functional
neuroimaging34 (34) or ApoE genotyping may improve the ability of any of
these measures alone to predict cognitive decline.35

In vivo imaging of amyloid plaques and neurofibrillary tangles

This is the most innovative structural imaging approach currently under
development. Although not yet widely available, the technology offers con-
DETECTING ALZHEIMER’S DISEASE AT THE PRE-SYMPTOMATIC STAGES 263

siderable promise in studies of new drug development and eventually in dif-

ferential diagnosis and early detection of dementia. The evidence for neuritic
plaque (NP) and neurofibrillary tangle (NFT) accumulation years prior to
clinical AD diagnosis suggests that in vivo methods that directly image these
pathognomic lesions would be useful presymptomatic detection technologies.
Current methods for measuring brain amyloid, such as histochemical stains,
require tissue fixation on post-mortem or biopsy material. Available in vivo
methods for measuring NPs or NFTs are indirect (e.g., cerebrospinal fluid
measures).36 Studies that may lead to direct in vivo human A imaging include
various radio-labelled probes using small organic and organometallic mol-
ecules capable of detecting differences in amyloid fibril structure or amyloid
protein sequences.37 Investigators also have used chrysamine-G, a carboxylic
acid analogue of Congo red; an amyloid-staining histologic dye; 38 serum
amyloid P component, a normal plasma glycoprotein that binds to amyloid
deposit fibrils;39 or monoclonal antibodies.40 Methodological difficulties that
hinder progress with these techniques include poor blood–brain barrier cross-
ing and limited specificity and sensitivity. In addition, most approaches do
not measure both NPs and NFTs.
In a recent study, Barrio et al. 41 used a hydrophobic radiofluorinated
derivative of 1,1-dicyano-2-[6-(dimethylamino)naphthalen-2-yl]propene
(FDDNP)42 with PET to measure the cerebral localization and load of NFTs
and SPs in AD patients (N=7) and controls (N=3). The FDDNP was injected
intravenously and found to readily cross the blood-brain barrier in propor-
tion to blood flow, as expected from highly hydrophobic compounds with
high membrane permeability. Greater accumulation and slower clearance
of FDDNP was observed in brain regions with high concentrations of NPs
and NFTs, particularly the hippocampus, amygdala, and entorhinal cortex.
The FDDNP residence time in these regions showed significant correlations
with immediate and delayed memory performance measures,43 and areas of
low glucose metabolism correlated with high FDDNP activity retention. The
probe showed visualization of NFTs, NPs and diffuse amyloid in AD brain
specimens using in vitro fluorescence microscropy, which matched results
using conventional stains (e.g. thioflavin S) in the same tissue specimens.
Thus, FDDNP-PET imaging is a promising non-invasive approach to longi-
tudinal evaluation of NP and NFT deposition in preclinical AD.

Magnetic resonance spectroscopy

This approach is another innovative technique that is only recently been
studied in the context of dementia. Initial studies of MRS as a preclinical
AD detection technique found significantly lower NAA concentrations in
AD and AAMI subjects compared with controls.44 Mean inositol concentra-
tion was significantly higher in AD than in controls, whereas AAMI subjects
had intermediate values. Another study focused on patients with Down’s
syndrome because they invariably develop Alzheimer-type pathology by the
time they reach their thirties or forties. Concentrations of myoinositol and
264 THE AGEING BRAIN

choline-containing compounds using 1H MRS were significantly higher in

the occipital and parietal regions in 19 non-demented adults with Down’s
syndrome and 17 age- and sex-matched healthy controls.45 Moreover, older
Down’s syndrome subjects (42–62 years) had higher myo-inositol levels than
younger subjects (28–39 years) suggesting that this approach may eventually
be useful as a preclinical AD marker.

Functional Imaging

Positron emission tomography

In recent years, clinicians have shown greater interest in the use of PET
scanning for dementia diagnosis as access to PET centers has increased and
mounting evidence of its diagnostic accuracy has come to light. Using fluoro-
deoxyglucose PET (FDG-PET), our group reported that parietal hypome-
tabolism predicted future AD in people with questionable dementia46 and
that even people with very mild age-related memory complaints have baseline
PET patterns predicting cognitive decline after three years.47 These initial
studies using PET for early AD detection emphasized family history of AD
as a risk factor for future cognitive decline. A change in focus came with the
discovery of the ApoE genetic risk for AD. The first report combining PET
imaging and ApoE genetic risk in people with a family history of AD included
12 non-demented relatives with ApoE ε4, 19 relatives without ApoE ε4, and
compared them to seven probable AD patients.14 “At-risk” subjects had mild
memory complaints, normal cognitive performance, and at least two relatives
with AD. Subjects with ApoE ε4 did not differ from those without ApoE ε4 in
mean age (56.4 vs. 55.5 years) or in neuropsychological performance. Parietal
metabolism was significantly lower and left-right parietal asymmetry higher
in at-risk subjects with ApoE ε4 compared to those without ApoE ε4. Patients
with dementia had significantly lower parietal metabolism than did at-risk
subjects with ApoE ε4.
The following year, Reiman et al.6 replicated these results and extended
them to other brain regions. They found hypometabolism in temporal, pre-
frontal and posterior cingulate regions in a study of 11 non-demented ApoE
ε4 homozygotes (4/4 genotype) and 22 ApoE ε3 homozygotes (3/3 genotype)
of similar ages, i.e. in their mid-fifties, to those in our own initial study. They
also applied an automated image analysis method, wherein metabolic reduc-
tions were standardized using three-dimensional stereotactic surface projec-
tions from FDG PET scans of AD patients compared with controls.48 The
results from these two studies6,14 provided independent confirmation of an
association between genetic risk and regional cerebral glucose hypometabo-
lism.
Our group recently confirmed these two initial reports in a study that
included none of the subjects participating in our previous report on ApoE
and PET.14 We studied 65 subjects in the 50 to 84 year age range (mean+SD
DETECTING ALZHEIMER’S DISEASE AT THE PRE-SYMPTOMATIC STAGES 265

= 67.3+9.4 years), with or without a family history of AD.49 Of the 65 sub-

jects, 54 were non-demented (27 ApoE ε4 carriers and 27 were subjects with-
out ApoE ε4), and 11 were demented and diagnosed with probable AD.49
The non-demented subjects were aware of a gradual onset of mild memory
complaints (e.g. misplacing familiar objects, difficulty remembering names)
but had memory performance scores within the norms for cognitively intact
persons of the same age and educational level. The ApoE ε4 carriers had a
small and non-significant but consistent reduction in cognitive performance.
As predicted, baseline comparisons among the three subject groups indicated
the lowest metabolic rates for the AD group, intermediate rates for the non-
demented ApoE ε4 carriers, and highest rates for the non-demented group
without ApoE ε4 in several cortical regions, including inferior parietal, lateral
temporal, and posterior cingulate (Figure 1).
Additional studies using FDG-PET have focused on older patients with
Down’s syndrome who are at risk for AD.50 The investigators hypothesized
that an audiovisual stimulation paradigm would serve as a stress test and
reveal abnormalities in parietal and temporal cerebral glucose metabolism
before dementia developed. At mental rest, younger and older patients with
Down’s syndrome did not differ in glucose metabolic patterns. During audio-
visual stimulation, however, the older patients showed significantly lower
parietal and temporal metabolism. Families with familial AD linked to chro-
mosome 14 or APP mutations have been studied with FDG-PET as well.51 In
such families with early-onset AD, approximately half of relatives who live
to the age at risk will develop AD. While pedigree members with AD show
typical parietal and temporal hypometabolism, asymptomatic relatives at risk
for AD show a similar but less severe hypometabolic pattern.

Single photon emission computed tomography

Johnson et al.52 used SPECT with a 99mTc-HMPAO to study longitudinal
cerebral perfusion of patients with questionable AD (Clinical Dementia Rat-
ing [CDR] = 0.5)53 and controls. Regional decreases in perfusion in patients
whose diagnosis converted to AD were most prominent in the hippocampal-
amygdaloid complex, the anterior and posterior cingulate, and the anterior
thalamus. Including ApoE status did not influence results. A direct com-
parison of FDG-PET and HMPAO-SPECT in their ability to differentiate AD
from vascular dementia indicated higher diagnostic accuracy for PET regard-
less of dementia severity.54 Using ROC curves, PET diagnostic accuracy was
better than SPECT for MMSE > 20 (87.2% vs. 62.9%) and for MMSE <
20 (100% vs. 81.2%). Other studies have confirmed a lower sensitivity for
even high-resolution SPECT compared with PET.55 Moreover, the parietal
hypoperfusion observed using SPECT in AD patients has been observed in
such other conditions as normal ageing, vascular dementia, post-hypoxic
dementia, and sleep apnea.56
266 THE AGEING BRAIN

Figure 1. Examples of PET images (comparable parietal lobe levels) co-registered

to each subject’s baseline MRI scan for an 81-year-old non-demented
woman (ApoE 3/3 genotype; upper images), a 76-year-old non-demented
woman (ApoE 3/4 genotype; middle images), and 79-year-old woman with
AD (ApoE 3/4 genotype; lower images). The last column shows two-year
follow-up scans for the non-demented women. Compared with the non-
demented subject without ApoE ε4, the non-demented ApoE ε4 carrier had
18% (right) and 12% (left) lower inferior parietal cortical metabolism,
while the demented woman’s parietal cortical metabolism was 20% (right)
and 22% (left) lower, as well as more widespread metabolic dysfunction
due to disease progression. Two-year follow-up scans showed minimal
parietal cortical decline for the woman without ApoE ε4, but bilateral
parietal cortical decline for the non-demented woman with ApoE ε4, who
also met clinical criteria for mild AD at follow-up. MRI scans were within
normal limits.49

Functional MRI
Two recent studies have combined ApoE genotyping and fMRI in persons
at risk for AD. Bookheimer at al.57 performed fMRI studies while 30 cogni-
tively intact middle-aged and older persons (mean age 63 years) memorized
and retrieved unrelated word pairs. The 16 ApoE ε4 carriers did not differ
DETECTING ALZHEIMER’S DISEASE AT THE PRE-SYMPTOMATIC STAGES 267

Figure 2. These statistical parametric maps of recall vs. control blocks for ApoE ε4
carriers and non-carriers were standardized into a common coordinate
system. Both groups showed significant MRI signal intensity increases in
frontal, temporal and parietal regions, and the ApoE ε4 group had greater
extent and intensity of activation. The ApoE ε4 group showed additional
activation in the left parahippocampal region, left dorsal prefrontal cortex,
and other regions in the inferior and superior parietal lobes, and anterior
cingulate.57

significantly from the 14 subjects without ApoE ε4 in age, prior educational

achievement, or rates of AD family history. Brain activation patterns were
determined during both learning and retrieval task periods and analyzed using
between-group and within-subject approaches. Memory performance was
reassessed on 12 subjects after two years of follow-up. The ApoE ε4 carriers
had significantly greater magnitude and spatial extent of MRI signal intensity
during memory performance in regions affected by AD, including bilateral
hippocampal and left parietal and prefrontal regions (Figure 2). This pattern
of activation was greater in the left hemisphere, consistent with the verbal
nature of the task, and during the retrieval rather than the learning condition.
Longitudinal data indicated that greater baseline brain activation correlated
with verbal memory decline assessed two years later. The greater signal in
268 THE AGEING BRAIN

subjects with the ApoE ε4 genetic risk suggests that the brain may recruit
additional neurons to compensate for subtle deficits. Moreover, the longitu-
dinal data are encouraging that functional MRI may be a useful approach to
prediction of future cognitive decline and early AD detection.
By contrast, other kinds of memory tasks may produce different patterns
of brain activation. In another study of persons at risk for AD, visual naming
and letter fluency tasks were used to activate brain areas involved in object
and face recognition during functional MRI scanning.58 Subjects in the high-
risk group had at least one first-degree relative with AD and one ApoE ε4
allele. The low risk group was matched for age, education, and cognitive
performance. The high-risk group showed reduced activation in the mid-
and posterior inferotemporal regions bilaterally. Such decreased activation
patterns could result from subclinical neuropathology in the inferotemporal
region or in the inputs to that region.

Longitudinal studies of glucose metabolism of persons at risk for

dementia
Two research groups — UCLA and University of Arizona — have reported
their longitudinal FDG-PET follow-up data on non-demented persons at risk
for AD. At UCLA, a total of 20 non-demented subjects (10 ApoE ε4 carriers
and 10 without ApoE ε4) have received repeat PET and neuropsychologi-
cal testing two years after baseline assessment (mean+SD for follow-up was
27.9+1.7 months).49 The 10 ApoE ε4 carriers available for longitudinal study
were similar to the 10 non-carriers in mean+SD age (67.9+8.9 vs. 69.6+8.1
years) and educational achievement (14.4+1.8 vs. 16.4+2.8 years). Memory
performance scores did not differ significantly according to genetic risk either
at baseline or follow-up and the ApoE ε4 carriers and non-carriers did not
differ significantly in cognitive change after two years.
The ROI analysis of PET scans performed after two years showed signifi-
cant glucose metabolic decline (4%) in the left posterior cingulate region in
ApoE ε4 carriers. The SPM analysis showed significant metabolic decline in
the inferior parietal and lateral temporal cortices with the greatest magnitude
(5%) of metabolic decline in the temporal cortex (Figure 3). After correction
for multiple comparisons, this decline remained significant for the ApoE ε4
group, wherein a decrease in metabolism was documented for every subject.
Based upon these data from only 10 subjects, the estimated power of PET
under the most conservative scenario is 0.9 to detect a 1-unit decline from
baseline to follow-up using a one-tailed test. Such findings suggest that
combining PET and AD genetic risk measures will allow investigators to
use relatively small sample sizes when testing anti-dementia treatments in
preclinical AD stages. The University of Arizona group also found that ApoE
ε4 heterozygotes have significant two-year declines in regional brain activity,
the largest of which is in temporal cortex, and that these reductions are sig-
nificantly greater than those in ApoE ε4 non-carriers. Their findings suggest
that as few as 22 cognitively normal, middle-aged ApoE ε4 heterozygotes
DETECTING ALZHEIMER’S DISEASE AT THE PRE-SYMPTOMATIC STAGES 269

Figure 3. Regions showing the greatest metabolic decline after two years of longi-
tudinal follow-up in non-demented subjects with ApoE ε4 (SPM analysis)
included the right lateral temporal and inferior parietal cortex (brain on the
left side of figure). Voxels undergoing metabolic decline (p<0.001, before
correction) are displayed in color, with peak significance (z=4.35) occurring
in Brodmann’s area 21 of the right middle temporal gyrus.49

would be needed in each treatment arm (i.e. active drug and placebo) to test
a prevention therapy over a two-year period.59

Clinical trials of pre-symptomatic patients using neuroimaging surrogate

markers
The longitudinal findings of significant parietal and temporal metabolic
decline in asymptomatic persons at risk for AD because of age or genetic risk
or both have now been confirmed at two centers in separate subject cohorts.
Together these studies indicate that combining PET imaging of glucose
metabolism and genetic risk may be useful outcome markers in AD prevention
trials. Functional brain imaging techniques could be used to track pre-clinical
cognitive decline and test candidate prevention therapies without having to
perform prolonged multi-site studies using incipient AD as the primary out-
come measure. The consistency and extent of the metabolic decline in these
well-screened populations indicate that the PET measures provide adequate
power to observe such decline in relatively small subject groups. A similar but
less striking metabolic decline pattern was noted in subjects without ApoE ε4
such that larger groups per treatment arm would be needed.
These observations provide an opportunity for pre-symptomatic treatment
trials not previously available. Until now, such trials involved studies of pre-
clinical subjects with more severe memory impairments consistent with MCI,
wherein approximately 50% of subjects actually develop dementia over a 3–4
year period. The MCI trials have required hundreds of subjects for adequate
power. These trials use a categorical variable, incipient dementia, as the pri-
270 THE AGEING BRAIN

mary outcome measure. The introduction of FDG-PET imaging combined

with ApoE ε4 genetic risk increases efficiency and reduces costs by addressing
the research questions with fewer subjects. Our group is currently performing
two such placebo-controlled trials, one using the cyclooxygenase-2 inhibitor
celecoxib and the other using the cholinesterase inhibitor donepezil.

Acknowledgements

Supported in part by the Alzheimer’s Association; the Fran and Ray Stark
Foundation Fund for Alzheimer’s Disease Research, Los Angeles, Calif; and
NIH grants MH52453, AG10123, and AG13308. The views expressed are
those of the author and do not necessarily represent those of the Department
of Veterans Affairs.

References

1. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical

diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group
under the auspices of Department of Health and Human Services Task Force on
Alzheimer’s Disease. Neurology. 1984; 34:939–944.
2. Crook T, Bartus RT, Ferris SH, Whitehouse P, Cohen GD, Gershon S. Age-associ-
ated memory impairment: proposed diagnostic criteria and measures of clinical
change — Report of a National Institute of Mental Health Work Group. Dev
Neuropsychol. 1986; 2:261–276.
3. Larrabee GJ, Crook TH. Estimated prevalence of age-associated memory impair-
ment derived from standardized tests of memory function. Int Psychogeriatr.
1994; 6:95–104.
4. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild
cognitive impairment: clinical characterization and outcome. Arch Neurol. 1999;
56:303–308.
5. Andersen K, Nielsen H, Lolk A, Andersen J, Becker I, Kragh-Sørensen P. Inci-
dence of very mild to severe dementia and Alzheimer’s disease in Denmark: the
Odense Study. Neurology. 1999; 52:85–90.
6. Reiman EM, Caselli RJ, Yun LS, Chen K, Bandy D, Minoshima S, Thibodeau SN,
Osborne D. Preclinical evidence of Alzheimer’s disease in persons homozygous
for the epsilon 4 allele for apolipoprotein E [see comments]. N Engl J Med. 1996;
334:752–758.
7. Price JL, Morris JC. Tangles and plaques in nondemented aging and “preclinical”
Alzheimer’s disease. Ann Neurol. 1999; 45:358–368.
8. Braak H, Braak E. Neuropathological stageing of Alzheimer-related changes. Acta
Neuropathol. 1991; 82:239–259.
9. Arai T, Ikeda K, Akiyama H, Haga C, Usami M, Sahara N, Iritani S, Mori H. A
high incidence of apolipoprotein E epsilon4 allele in middle-aged non-demented
subjects with cerebral amyloid beta protein deposits. Acta Neuropathol. 1999;
97:82–84.
10. Snowdon DA, Kemper SJ, Mortimer JA, Greiner LH, Wekstein DR, Markesbery
WR. Linguistic ability in early life and cognitive function and Alzheimer’s dis-
ease in late life. Findings from the Nun Study. J Amer Med Assoc. 1996; 275:
528–532.
DETECTING ALZHEIMER’S DISEASE AT THE PRE-SYMPTOMATIC STAGES 271

11. Beach TG, Honer WG, Hughes LH. Cholinergic fibre loss associated with diffuse
plaques in the non-demented elderly: the preclinical stage of Alzheimer’s disease?
Acta Neuropathol. 1997; 93:146–153.
12. Shen J, Barnes CA. Age-related decrease in cholinergic synaptic transmission in
three hippocampal subfields. Neurobiol Aging. 1996; 17:439–451.
13. Small GW, Ercoli LM, Huang S-C, Komo S, Bookheimer SY, Saxena S, Silverman
DHS, Mega MS, Mazziotta JC, Wu HM, Cummings JL, Phelps ME. PET and
genetic risk for Alzheimer disease. J Nucl Med. 1999;40 (Suppl.):70.
14. Small GW, Mazziotta JC, Collins MT, Baxter LR, Phelps ME, Mandelkern MA,
Kaplan A, La Rue A, Adamson CF, Chang L. Apolipoprotein E type 4 allele and
cerebral glucose metabolism in relatives at risk for familial Alzheimer disease.
J Amer Med Assoc. 1995; 273:942–947.
15. de Leon MJ, George AE, Golomb JC, Convit A, Kluger A, De Santi S, McRae
T, Ferris SH, Reisberg B, Ince C, Rusinek H, Bobinski M, Quinn B, Miller DC,
Wisniewski HM. Frequency of hippocampal formation atrophy in normal aging
and Alzheimer’s disease. Neurobiol Aging. 1997; 18:1–11.
16. Boone KB, Miller BL, Lesser IM, Mehringer CM, Hill-Gutierrez E, Goldberg MA,
Berman NG. Neuropsychological correlates of white-matter lesions in healthy
elderly subjects. A threshold effect. Arch Neurol. 1992; 49:549–554.
17. Lopez OL, Becker JT, D R, Wess J, Boller F, Reynolds CF 3d, Panisset M.
Neuropsychiatric correlates of cerebral white-matter radiolucencies in probable
Alzheimer’s disease. Arch Neurol. 1992; 49:828–834.
18. Rusinek H, de Leon MJ, George AE, Stylopoulos LA, Chandra R, Smith G, Rand
T, Mourino M, Kowalski H. Alzheimer disease: measuring loss of cerebral gray
matter with MR imaging. Neuroradiology. 1991; 178:109–114.
19. Kesslak JP, Nalcioglu O, Cotman CW. Quantification of magnetic resonance
scans for hippocampal and parahippocampal atrophy in Alzheimer’s disease.
Neurology. 1991; 41:51–54.
20. Pearlson GD, Harris GJ, Powers RE, Barta PE, Camargo EE, Chase GA, Noga
JT, Tune LE. Quantitative changes in mesial temporal volume, regional cerebral
blood flow, and cognition in Alzheimer’s disease. Arch Gen Psychiat. 1992; 49:
402–408.
21. Laakso MP, Soininen H, Partanen K, Lehtovirta M, Helkala EL, Hallikainen
M, Hanninen T, Vainio P, Soininen H. MRI of the hippocampus in Alzheimer’s
disease: sensitivity, specificity, and analysis of the incorrectly classified subjects.
Neurobiol Aging. 1998; 19:23–31.
22. Pasquier F, Lavenu I, Lebert F, Jacob B, Steinling M, Petit H. The use of SPECT
in a multidisciplinary memory clinic. Dement Geriatr Cogn. 1997; 8:85–91.
23. Pucci E, Belardinelli N, Regnicolo L, Nolfe G, Signorino M, Salvolini U, Ange-
leri F. Hippocampus and parahippocampal gyrus linear measurements based on
magnetic resonance in Alzheimer’s disease. Eur Neurol. 1998; 39:16–25.
24. Jack CR, Petersen RC, Xu YC, O’Brien PC, Smith GE, Ivnik RJ, Boeve BF, War-
ing SC, Tangalos EG, Kokmen E. Prediction of AD with MRI-based hippocampal
volume in mild cognitive impairment. Neurology. 1999; 52:1397–1403.
25. Wahlund L-O, Julin P, Lindqvist J, Scheltens P. Visual assessment of medial
temporal atrophy in demented and healthy control subjects: correlation with
volumetry. Psychiat Res Neuroim Section. 1999; 90 :193–199.
26. Killiany RJ, Moss MB, Albert MS, Sandor T, Tieman J, Jolesz F. Temporal lobe
regions on magnetic resonance imaging identify patients with early Alzheimer’s
disease. Arch Neurol. 1993; 50:949–954.
27. Laakso MP, Partanen K, Riekkinen P, Lehtovirta M, Helkala EL, Hallikainen M,
Hanninen T, Vainio P, Soininen H. Hippocampal volumes in Alzheimer’s disease,
Parkinson’s disease, with and without dementia, and in vascular dementia: An
MRI study. Neurology. 1996; 46:678–681.
272 THE AGEING BRAIN

28. Reiman EM, Uecker A, Caselli RJ, Lewis S, Bandy D, de Leon MJ, De Santi S,
Convit A, Osborne D, Weaver A, Thibodeau SN. Hippocampal volumes in cogni-
tively normal persons at genetic risk for Alzheimer’s disease. Ann Neurol. 1998;
44:288–291.
29. Erkinjuntti T, Gao F, Lee DH, Eliasziw M, Merskey H, Hachinski CV. Lack of
difference in brain hyperintensities between patients with early Alzheimer’s dis-
ease and control subjects. Arch Neurol. 1994; 51:260–268.
30. Mauri M, Sibilla L, Bono G, Carlesimo GA, Sinforiani E, Martelli A. The role of
morpho-volumetric and memory correlations in the diagnosis of early Alzheimer
dementia. J Neurol. 1998; 245:525–530.
31. Bobinski M, de Leon MJ, Convit A, De Santi S, Wegiel J, Tarshish CY, Saint
Louis LA, Wisniewski HM. MRI of entorhinal cortex in mild Alzheimer’s disease.
Lancet. 1999;353:38–40.
32. Geroldi C, Pihlajamaki M, Laakso MP, DeCarli C, Beltramello A, Bianchetti A,
Soininen H, Trabucchi M, Frisoni GB. APOE-ε4 is associated with less frontal and
more medial temporal lobe atrophy in AD. Neurology. 1999; 53:1825–1832.
33. Soininen H, Partanen K, Pitkanen A, Hallikainen M, Hänninen T, Helisalmi S,
Mannermaa A, Ryynänen M, Koivisto K, Riekkinen P. Decreased hippocampal
volume asymmetry on MRIs in nondemented elderly subjects carrying the apoli-
poprotein E ε4 allele. Neurology. 1995; 45:1467–1472.
34. Mattman A, Feldman H, Forster B, Li D, Szasz I, Beattie BL, Schulzer M. Regional
HmPAO SPECT and CT measurements in the diagnosis of Alzheimer’s disease.
Can J Neurol Sci. 1997; 24:22–28.
35. Jack CR, Petersen RC, Xu YC, O’Brien PC, Waring SC, Tangalos EG, Smith GE,
Ivnik RJ, Thibodeau SN, Kokmen E. Hippocampal atrophy and apolipoprotein
E genotype are independently associated with Alzheimer’s disease. Ann Neurol.
1998; 43:303–310.
36. Motter R, Vigo-Pelfrey C, Kholodenko D, Barbour R, Johnson-Wood K, Galasko
D, Chang L, Miller B, Clark C, Green R. Reduction of beta-amyloid peptide42 in
the cerebrospinal fluid of patients with Alzheimer’s disease. Ann Neurol. 1995;
38:643–648.
37. Ashburn TT, Han H, McGuinness BF, Lansbury PT. Amyloid probes based on
Congo Red distinguish between fibrils comprising different peptides. Chem Biol.
1996; 3:351–358.
38. Klunk WE, Debnath ML, Pettegrew JW. Chrysamine-G binding to Alzheimer and
control brain: autopsy study of a new amyloid probe. Neurobiol Aging. 1995; 16:
541–548.
39. Lovat LB, O’Brien AA, Armstrong SF, Madhoo S, Bulpitt CJ, Rossor MN, Pepys
MB, Hawkins PN. Scintigraphy with 123I-serum amyloid P component in Alzhe-
imer disease. Alz Dis Assoc Dis. 1998; 12:208–210.
40. Majocha RE, Reno JM, Friedland RP, VanHaight C, Lyle LR, Marotta CA.
Development of a monoclonal antibody specific for beta/A4 amyloid in Alzhei-
mer’s disease brain for application to in vivo imaging of amyloid angiopathy.
J Nucl Med. 1992; 33:2184–2189.
41. Barrio JR, Huang S-C, Cole GM, Satyamurthy N, Petric A, Small GW. PET imag-
ing of tangles and plaques in Alzheimer disease. J Nucl Med. 1999;40 (Suppl.):
70P–71P.
42. Jacobson A, Petric A, Hogenkamp D, Sinur A, Barrio JR. 1,1-dicyano-2-(6-
dimethylamino)naphthalen-2-yl)propene (DDNP): a solvent polarity and viscos-
ity sensitive fluorophore for fluorescence microscopy. J Amer Chem Soc. 1996;
118:5572–5579.
43. Wechsler D. Wechsler Memory Scale-Revised Manual. San Antonio: The Psycho-
logical Corp – Harcourt Brace Jovanovich; 1987.
DETECTING ALZHEIMER’S DISEASE AT THE PRE-SYMPTOMATIC STAGES 273

44. Parnetti L, Lowenthal DT, Presciutti O, Pelliccioli GP, Palumbo R, Gobbi G,

Chiarini P, Palumbo B, Tarducci R, Senin U. 1H-MRS, MRI-based hippocampal
volumetry, and 99mTc-HMPAO-SPECT in normal aging, age-associated memory
impairment, and probable Alzheimer’s disease. J Amer Geriatr Soc. 1996; 44:
133–138.
45. Huang W, Alexander GE, Daly EM, Shetty HU, Krasuski JS, Rapoport SI, Scha-
piro MB. High brain myo-inositol levels in the predementia phase of Alzheimer’s
disease in adults with Down’s syndrome: a 1H MRS study. Am J Psychiat. 1999;
156:1879–1886.
46. Kuhl DE, Small GW, Riege WH, Fujikawa DG, Metter EJ, Benson DF, Ashford
JW, Mazziotta JC, Maltese A, Dorsey DA. Cerebral metabolic patterns before
diagnosis of probable Alzheimer’s disease. J Cerebr Blood F Met. 1987; 7
(Suppl 1):S406.
47. Small GW, La Rue A, Komo S, Kaplan A, Mandelkern MA. Predictors of cogni-
tive change in middle-aged and older adults with memory loss. Am J Psychiat.
1995; 152:1757–1764.
48. Minoshima S, Frey KA, Koeppe RA, Foster NL, Kuhl DE. A diagnostic approach
in Alzheimer’s disease using three-dimensional stereotactic surface projections of
fluorine-18-FDG PET. J Nucl Med. 1995; 36:1238–1248.
49. Small GW, Ercoli LM, Silverman DHS, Huang S-C, Komo S, Bookheimer SY,
Lavretsky H, Miller K, Siddarth P, Mazziotta JC, Saxena S, Wu HM, Mega MS,
Cummings JL, Saunders AM, Pericak-Vance MA, Roses AD, Barrio JR, Phelps
ME. Cerebral metabolic and cognitive decline in persons at genetic risk for Alzhe-
imer’s disease. Proc Natl Acad Sci USA. 2000; 97:6037–6042.
50. Pietrini P, Dani A, Furey ML, Alexander GE, Freo U, Grady CL, Mentis MJ, Man-
got D, Simon EW, Horwitz B, Haxby JV, Schapiro MB. Low glucose metabolism
during brain stimulation in older Down’s syndrome subjects at risk for Alzheim-
er’s disease prior to dementia. Am J Psychiat. 1997; 154:1063–1069.
51. Rossor MN, Kennedy AM, Frackowiak RS. Clinical and neuroimaging features
of familial Alzheimer’s disease. Ann NY Acad Sci. 1996; 777:49–56.
52. Johnson KA, Jones K, Holman BL, Becker JA, Spiers PA, Satlin A, Albert MS.
Preclinical prediction of Alzheimer’s disease using SPECT. Neurology. 1998; 50:
1563–1571.
53. Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL. A new clinical scale for
the staging of dementia. Br J Psychiat. 1982; 140:566–572.
54. Mielke R, Heiss W-D. Positron emission tomography for diagnosis of Alzheimer’s
disease and vascular dementia. J Neural Transm. 1998; 53 (Suppl):237–250.
55. Messa C, Perani D, Lucignani G, Zenorini A, Zito F, Rizzo G, Grassi F, Del Sole
A, Franceschi M, Gilardi MC. High-resolution technetium-99m-HMPAO SPECT
in patients with probable Alzheimer’s disease: comparison with fluorine-18-FDG
PET. J Nucl Med. 1994; 35:210–216.
56. Miller BL, Mena S, Daly J, Giombetti RJ, Goldbergt MA, Lesser I, Garrett K, Vil-
lanueva-Meyer J, Liu C-K. Temporal-parietal hypoperfusion with single-photon
emission computerized tomography in conditions other than Alzheimer’s disease.
Dementia. 1990; 1:41–45.
57. Bookheimer SY, Strojwas MH, Cohen MS, Saunders AM, Pericak-Vance MA,
Mazziotta JC, Small GW. Brain activation in older people at genetic risk for
Alzheimer’s disease. N Engl J Med. 2000; 343:450–456.
58. Smith CD, Andersen AH, Kryscio RJ, Schmitt FA, Kindy MS, Blonder LX, Avi-
son MJ. Altered brain activation in cognitively intact individuals at high risk for
Alzhiemer’s disease. Neurology. 1999; 53:1391–1396.
59. Reiman EM, Uecker A, Gonzalez-Lima F, Minear D, Chen K, Callaway NL,
Berndt JD, Games D. Tracking Alzheimer’s disease in transgenic mice using
fluorodeoxyglucose autoradiography. NeuroReport. 2000 11:987–991.
Chapter 16

PARKINSONISM AND
AGEING
John GL Morris*, Mariese A Hely, and
Glenda M Halliday

Introduction

It is a matter of everyday experience that, in the journey from childhood to

old age, the way in which we move, speak, hold ourselves and walk gradually
changes. A movement or gesture by a skilled actor can, in a moment, leave his
audience in no doubt as to the age of the person being depicted; our posture
and movements denote our age. The changes from childhood to adulthood
reflect maturation of the brain and body. But what determines the changes
from middle to old age? Is this normal “wear and tear” of the bones, joints
and muscles, or are the changes we see due to disease? Friends who have not
seen a patient with early Parkinsonism for a while may be struck at how they
have “aged”: they have become stooped and slow. Can Parkinsonism be con-
sidered a form of accelerated ageing or, conversely, how much of “normal”
ageing is due to incipient Parkinson’s disease? In this chapter, we compare
and contrast ageing and Parkinsonism.

Posture

Perhaps the most telling sign of advancing years is the emergence of a stoop.
The spine becomes flexed, the head drops forward. Parkinson’s disease (PD)
also causes flexion of the neck and spine but here, the elbows, knees and hips
are often also flexed causing the so-called simian (“ape-like”) posture. The

*To whom correspondence should be addressed.

276 THE AGEING BRAIN

flexed posture of Parkinson’s disease reflects the distribution of tone within

the flexor and extensor muscles of the spine and limbs*.
The question arises: is the posture of old age due to factors such as wedg-
ing of the vertebrae due to osteoporosis, reduced mobility associated with
osteo-arthritis and muscle weakness and wasting secondary to disuse? Or are
there also changes in the brain which contribute to this? With advancing age
there is progressive atrophy of the brain1 due to shrinkage2 of neurones and
supporting cells but the association between these changes in the brain and
posture is uncertain. Lesions of the white matter of the cerebral hemispheres
become increasingly common with advancing age. These are due to lacunes,
perivascular spaces and gliosis probably secondary to infarction.3 Flexion of
the trunk is unlikely to be due to such vascular disease for in patients with
severe subcortical arteriosclerotic encephalopathy (Binswanger’s disease), the
posture is upright.4
The underlying abnormality in Parkinson’s disease is degeneration of the
dopaminergic nigro-striatal tract. Unilateral nigrostriatal lesions in animal
models of Parkinson’s disease produce postural abnormalities.5 Yet treatment
with the precursor of dopamine, levodopa, while improving most aspects of
the disease, does not usually improve posture to a great extent. This suggests
that the abnormality of posture in Parkinson’s disease results from involve-
ment of non-dopaminergic systems. More recently, the combination of
levodopa and bilateral subthalamic stimulation6 has been shown to improve
posture. The mechanism of this aspect of the disease remains uncertain.

Gait and Balance

The gait of the very old lacks the speed, confidence and spring of the young.
In many older people, the gait also appears to reflect a fear of falling. It is
slower, the stride length is shortened, the base widened a little and turning is
executed in several steps rather than in a single smooth movement. This is the
so-called “cautious” gait7 which we all adopt on a ship’s deck in rough con-
ditions or when walking on slippery ice. Factors which contribute to such a
gait in the elderly include impaired eyesight (cataract, macular degeneration)
and osteo-arthritis, especially of the knees and hips. Sudden pain may cause
the leg to give way resulting in a fall. Painful arthritis may further modify
the gait producing the so-called “antalgic” or hobbling gait, where the nor-
mal leg swings faster than the painful leg, thereby minimising the time that
the painful limb has to bear weight. Many older patients have or have had
diseases which impair other principal components necessary for normal gait

* The posture in some of the variants of Parkinson’s disease is rather different. In

anterocollis, a feature of multiple system atrophy and sometimes dementia with
Lewy bodies, the overall posture of the trunk is upright but the neck is markedly
flexed. The posture of the trunk usually remains upright in progressive supranu-
clear palsy.
PARKINSONISM AND AGEING 277

and balance: the vestibular system (vestibular neuronitis, Meniere’s disease,

antibiotic ototoxicity), proprioception (diabetic and other forms of peripheral
neuropathy8), cerebellum (alcohol) and pyramidal system (cervical myelopa-
thy, stroke). Apart from diseases, there are age related changes in measures
such as peripheral nerve conduction velocity,9 and vestibular function10
which may have a bearing on movement and balance. The cause of changes
in gait with ageing may thus be seen as resulting from the cumulative effect
of minor impairment in a number of modalities, each of which is insufficient,
on its own, to impair gait.
An association has been found between impairment in balance in older
people and the findings of white matter lesions in the cerebral hemispheres11
and frontal atrophy12 on MRI scanning. Many of these patients have the so-
called marche a petit pas with slow, shuffling steps, a broad-based stance,
difficulty initiating gait (“start-hesitation”) but with well preserved hand
function, arm swing and facial expression and usually no festination (see
below)4. This “Parkinsonian-ataxic gait”4 is also seen in patients with low
pressure hydrocephalus and frontal tumours. Thompson and Marsden4 have
pointed out that the thalamo-cortical fibres destined for the supplementary
motor area (SMA) from the basal ganglia, and the thalamo-cortical fibres
from the leg area of the cerebellum destined for the motor cortex, are situ-
ated in the periventricular white matter. Fibres destined for the upper limbs
are more superficial, run a shorter course and may thus be less susceptible to
disease in the deep white matter.11,13
The gait of Parkinson’s disease is rather different. The earliest feature is
loss of arm swing, usually on only one side to begin with. Later, the stride
shortens and the posture becomes flexed. As the righting reflex is lost, falls
occur but this does not occur in idiopathic Parkinson’s disease in the early
years*. Notably, however bad the balance becomes, the gait remains narrow
based in Parkinson’s disease. A particular feature of the Parkinsonian gait
in advanced disease is its hurrying (“festinant”) quality where small steps
are taken rapidly as if the feet are trying to catch up with the stooped trunk
preceding them. In its most marked form, patients break involuntarily into a
run (“propulsion”), continuing until they fall or bump into something. A firm
pull from behind will cause such a patient to run backwards (“retropulsion”).
Associated with festination is “freezing” where the gait is suddenly arrested
often as the patient negotiates an obstacle such as a doorway, sometimes
the patient appears to “run on the spot” (“clutch slip”). The start-hesitation
and freezing of Parkinson’s disease may be overcome by visual cues such as
lines on the floor and steps; by contrast freezing in marche a petit pas is not
influenced by such cues.4,13 The gait in other patients with Parkinsonism is
characterised by extreme slowness and stiffness. What determines why one
patient hurries while another is slow is poorly understood.

* By contrast, falls occur early in the variants of Parkinson’s disease such as multiple
system atrophy and progressive supranuclear palsy.
278 THE AGEING BRAIN

The hallmark of the gait disturbance in Parkinson’s disease is the failure

to generate an appropriate stride length.14 This is thought to result from
disruption of the normal interaction between the basal ganglia and the SMA,
as a result of loss of the dopaminergic neurones of the nigro-striatal tract.15
Gait is improved by visual cues and made worse by distraction, something
that has also been observed in healthy elderly subjects.16 Gait in idiopathic
Parkinson’s disease is improved by levodopa therapy: distances are covered
more quickly, arm swing increases, and stride lengthens. By contrast, balance
is not improved and, paradoxically, patients may fall more frequently as their
mobility is enhanced by the effect of levodopa. The pathophysiology of loss of
balance in Parkinson’s disease is not well understood. It was associated with
frontal atrophy in one study of non-demented parkinsonian patients.17 Atten-
tion is now being directed to other structures such as the pedunculopontine
nucleus18 in an attempt to understand these aspects of Parkinson’s disease
which are unresponsive to levodopa therapy. Freezing often also persists
while levodopa is working. A surprising recent finding was that deprenyl, a
selective monoamine oxidase ‘B’ inhibitor which otherwise causes only minor
improvement in Parkinson’s disease, may help to prevent freezing;19 time will
tell if this finding holds up in clinical practice.

Eye Movements

Some restriction of upward gaze is apparent in many elderly otherwise nor-

mal patients.20 The basis of this is not known though it may be speculated
that it reflects ischaemia or degeneration of vertical gaze pathways in the
rostral brainstem. In Parkinson’s disease, there is reduction in the amplitude
of voluntary saccades, comparable to the shortening of stride length in gait.
The velocity of the saccades is not affected*. When asked to look from one
object to another, the patient’s eyes are seen to move in a series of small
“bunny hops” rather than in one clean leap. Levodopa does not improve this
abnormality but high frequency stimulation of the subthalamic nucleus has
been shown to improve some aspects of saccadic function.21 Smooth pursuit
is commonly impaired in both Parkinson’s disease and the elderly.

Speech

As people get older their speech changes. Often this can be attributed to ill-
fitting dentures causing distortion of consonants. More saliva is retained in
the mouth due, presumably, to reduced frequency of swallowing. There is also
a change in the timbre of the voice and some loss of volume. Loss of volume

* By contrast, in progressive supranuclear palsy, the most striking finding is of slow-

ing of saccades, and, later, impairment of downward gaze.
PARKINSONISM AND AGEING 279

(hypophonia) is the commonest speech abnormality in Parkinson’s disease

and may be attributed to reduced amplitude of movement (or akinesia) of the
respiratory muscles. More characteristic of Parkinson’s disease is the loss of
normal fluctuations of pitch and volume thereby producing a monotonous
voice. Like the gait, speech may be hurried, one sound running into the next
and with pauses where the power of speech is momentarily lost. Sounds may
be repeated, causing a stammer known as pallilalia. The pathological basis
of these problems in speech in Parkinson’s disease is uncertain. Speech is not
usually improved by levodopa therapy. Pallidotomy significantly worsens
speech abnormalities in some patients with Parkinson’s disease,22,23 though
this probably varies with the site and size of the lesion. Improvement in the
tempo and timbre of speech has been reported with subthalamic stimula-
tion.24

Facial Expression

The lines on our face deepen as we get older yet our faces lose some of the
animation of our youth. This is rarely marked and probably reflects changes
in mood and energy rather than any disorder of our motor system. By con-
trast, loss of facial expression in Parkinsonism may be profound. Unlike the
actor of whom it was said that

“Changes in mood crossed his face like small weather fronts”

(Daily Telegraph critic on Warren Clarke’s performance in A Respect-
able Trade on BBC1)

these patients’ faces are frozen and unblinking, apparently impervious to the
feelings of their owners until, on occasions, a feeling is strong enough to break
the ice and a smile slowly dawns. These features reflect akinesia and rigid-
ity and often improve with levodopa only to be replaced in some cases with
writhing of the facial features due to levodopa induced dyskinesia.

Alternating Fine Movements

While some loss of manual dexterity is measurable in normal patients as they

get older,25 this is not usually detectable by routine clinical testing. Every
patient with Parkinsonism has difficulty making alternating, piano-playing
movements of the index and middle fingers and most have difficulty toe tap-
ping. As previously discussed, the amplitude of most movements is reduced,
usually lessening further as the task continues. This feature of akinesia is also
seen in writing which becomes smaller as it goes across the page. Levodopa
markedly improves manual dexterity in idiopathic Parkinson’s disease. Simi-
lar improvement is seen with subthalamic stimulation.
280 THE AGEING BRAIN

Muscle Tone

This is often hard to test in elderly patients for it depends on their ability to
voluntarily relax and not actively assist or resist what the examiner is doing to
them. Asking such a patient to relax their legs induces a state of extreme stiff-
ness: the harder the examiner pulls, the more the patient resists. Sometimes
the problem is due to tenderness of the skin or muscles or painful arthritis,
and re-positioning of the examiner’s hands may help. This “voluntary” con-
traction of the muscles is sometimes called paratonia or gegenhalten. It is
commonly seen in the setting of advanced dementia where the patient resumes
the flexed posture of the foetus; in time, with contracture of the muscles, the
posture becomes permanent.

Rigidity

Parkinsonian rigidity is rather different. Here, the limbs can usually be moved
through their normal range but a resistance is felt which is constant through-
out that range (unlike spasticity, where a sudden increase in resistance or
“catch” may be felt, particularly if the examiner suddenly increases the rate
at which he moves the limb). Rigidity palpably increases in the upper limb,
when the patient is asked to slowly raise and lower the other arm at the shoul-
der (“tone reinforcement”). Rigidity, like akinesia, is improved by levodopa
therapy and is presumably a consequence of the loss of the dopaminergic
nigro-striatal neurones.

Resting Tremor

Many elderly patients have a low amplitude tremor of the outstretched hands
or occasional flurries of quivering of individual fingers. The presence of a
coarse resting tremor, usually in one hand more than the other, which per-
sists during walking and is temporarily abolished by voluntary movement of
the affected limb, usually signifies Parkinson’s disease and is not a feature of
normal ageing.26

Prevalence of Parkinsonism in the Elderly

As we age, our movements become slower. Bradykinesia without other signs

of Parkinsonism was commonly found in a study of normal subjects aged 75
years or more.20 Some features of Parkinsonism including slowness, stiffness,
gait disorder and tremor were found in 14.9% of people aged 65–74 years,
29.5% of people aged 75–84 years and 52.4% of people aged 85 or more in
a study of non-institutionalized elderly Boston residents.26 This is probably
PARKINSONISM AND AGEING 281

an overestimate for other studies have found a much lower prevalence of the
disease. In the Rotterdam study of 6969 subjects, each individually examined,
the prevalence of Parkinson’s disease was 1.0% for those aged 65–74, 3.1%
for those aged 75–84 and 4.3% for those aged 85–94 years.27 Similar findings
occurred in Sicilian and French community studies.28,29
In conclusion, though old age and Parkinson’s disease appear to share
many features, there are subtle differences between the two which point to
differing underlying mechanisms.1 The changes in Parkinson’s disease are
mainly due to degeneration of the nigro-striatal tracts. In old age, the changes
reflect a more generalized disturbance of neuronal function. Significantly,
the pathological changes in the substantia nigra in Parkinson’s disease, differ
from those due to ageing.30

References

1. Mahant PR, Stacy MA. Movement disorders and normal ageing. Neurol Clin.
2001; 19:553–563, vi.
2. Terry RD, DeTeresa R, Hansen LA. Neocortical cell counts in normal human
adult ageing. Ann Neurol. 1987; 21:530–539.
3. Baloh RW, Vintners HV. White matter lesions and disequilibrium in older people.
II Clinicopathologic correlation. Arch Neurol. 1995; 52:975–981.
4. Thompson PD, Marsden CD. Gait disorder of subcortical arteriosclerotic enceph-
alopathy: Binswanger’s disease. Movement Disord. 1987; 2:1–8.
5. Johnson RE, Schallert T, Becker JB. Akinesia and postural abnormality after
unilateral dopamine depletion. Behav Brain Res. 1999; 104:189–196.
6. Bejjani BP, Gervais D, Arnulf I, Papadopoulos S, Demeret S, Bonnet AM Cornu
P, Dmier P, Agid Y. Axial parkinsonian symptoms can be improved: the role of
levodopa and bilateral subthalamic stimulation. J Neurol Neurosur Ps. 2000; 68:
595–600.
7. Nutt JG, Marsden CD, Thompson PD. Human walking and higher-level gait
disorders, particularly in the elderly. Neurology. 1993; 43:268–279.
8. Bergin PS, Bronstein AM, Murray NM, Sancovic S, Zeppenfeld DK. Body sway
and vibration perception thresholds in normal ageing and in patients with
polyneuropathy. J Neurol Neurosur Ps. 1995; 58:335–340.
9. Rivner MH, Swift TR, Malik K. Influence of age and height on nerve conduction.
Muscle Nerve. 2001; 24:1134–1141.
10. Hajioff D, Barr-Hamilton RM, Colledge NR, Lewis SJ, Wilson JA. Re-evaluation
of normative electronystagmography data in healthy ageing. Clin Otolaryngol.
2000; 25:249–252.
11. Baloh RW, Yue Q, Socotch TM, Jacobson KM. White matter lesions and dis-
equilibrium in older people. I. Case-control comparison. Arch Neurol. 1995; 52:
970–974.
12. Kerber KA, Enrietto JA, Jacobson KM, Baloh RW. Disequilibrium in older peo-
ple: a prospective study. Neurology. 1998; 51:574–580.
13. Curran T, Lang AE. Parkinsonian syndromes associated with hydrocephalus: case
reports, a review of the literature, and pathophysiological hypotheses. Movement
Disord. 1994; 9:508–520.
14. Morris ME, Iansek R, Matyas TA, Summers JJ. Ability to modulate walking
cadence remains intact in Parkinson’s disease. J Neurol Neurosur Ps. 1994; 57:
1532–1534.
282 THE AGEING BRAIN

15. Morris ME, Iansek R, Matyas TA, Summers JJ. Stride length regulation in Parkin-
son’s disease. Normalization strategies and underlying mechanisms. Brain. 1996;
119:551–568.
16. Camicioli R, Howieson D, Lehman S, Kaye J. Talking while walking: the effect
of a dual task in aging and Alzheimer’s disease. Neurology. 1997; 48:955–958.
17. Durif F, Pollak P, Hommel M, Ardouin C, Le Bas JF, Crouzet C, Perret J. Rela-
tionship between levodopa-independent symptoms and central atrophy evaluated
by magnetic resonance imaging in Parkinson’s disease. Eur Neurol. 1992; 32:
32–36.
18. Pahapill PA, Lozano AM. The pedunculopontine nucleus and Parkinson’s disease.
Brain. 2000; 123:1767–1783.
19. Giladi N, McDermott MP, Fahn S, Przedborski S, Jankovic J, Stern M, Tanner C.
Freezing of gait in PD: prospective assessment in the DATATOP cohort. Neurol-
ogy. 2001; 56:1712–1721.
20. Waite LM, Broe GA, Creasey H, Grayson D, Edelbrock D, O’Toole B. Neurologi-
cal signs, aging, and the neurodegenerative syndromes. Arch Neurol. 1996; 53:
498–502.
21. Rivaud-Pechoux S, Vermersch AI, Gaymard B, Ploner CJ, Bejjani BP, Damier P,
Demeret S, Agid Y, Pierrot-Deseilligny C. Improvement of memory guided sac-
cades in parkinsonian patients by high frequency subthalamic nucleus stimula-
tion. J Neurol Neurosur Ps. 2000; 68:381–384.
22. Scott R, Gregory R, Hines N, Carroll C, Hyman N, Papanasstasiou V, Leather
C, Rowe J, Silburn P, Aziz T. Neuropsychological, neurological and functional
outcome following pallidotomy for Parkinson’s disease. A consecutive series of
eight simultaneous bilateral and twelve unilateral procedures. Brain. 1998; 121:
659–675.
23. Merello M, Starkstein S, Nouzeilles MI, Kuzis G, Leiguarda R. Bilateral palli-
dotomy for treatment of Parkinson’s disease induced corticobulbar syndrome and
psychic akinesia avoidable by globus pallidus lesion combined with contralateral
stimulation. J Neurol Neurosur Ps. 2001; 71:611–614.
24. Gentil M, Chauvin P, Pinto S, Pollak P, Benabid AL. Effect of bilateral stimula-
tion of the subthalamic nucleus on parkinsonian voice. Brain Lang. 2001; 78:
233–240.
25. Nutt JG, Lea ES, Van Houten L, Schuff RA, Sexton GJ. Determinants of tapping
speed in normal control subjects and subjects with Parkinson’s disease: differing
effects of brief and continued practice. Movement Disord. 2000; 15:843–849.
26. Bennett DA, Beckett LA, Murray AM, Shannon KM, Goetz CG, Pilgrim DM,
Evans DA. Prevalence of parkinsonian signs and associated mortality in a com-
munity population of older people. N Engl J Med. 1996; 334:71–76.
27. de Rijk MC, Breteler MM, Graveland GA, Ott A, Grobbee DE, van der Meche
FG, Hofman A. Prevalence of Parkinson’s disease in the elderly: the Rotterdam
Study. Neurology. 1995; 45:2143–2146.
28. Morgante L, Rocca WA, Di Rosa AE, De Domenico P, Grigoletto F, Meneghini
F, Reggio A, Savettieri G, Castiglione MB, Patti F. Prevalence of Parkinson’s
disease and other types of parkinsonism: a door-to-door survey in three Sicilian
municipalities. The Sicilian Neuro-Epidemiologic Study (SNES) Group. Neurol-
ogy. 1992; 42:1901–1907.
29. Tison F, Dartigues JF, Dubes L, Zuber M, Alperovitch A, Henry P. Prevalence of
Parkinson’s disease in the elderly: a population study in Gironde, France. Acta
Neurol Scand. 1994; 90:111–115.
30. Fearnley JM, Lees AJ. Ageing and Parkinson’s disease: substantia nigra regional
selectivity. Brain. 1991; 114:2283–2301.
Chapter 17

AGE VARIATION IN THE

PREVALENCE OF
DEPRESSION: ARE STUDY
FINDINGS MEANINGFUL?
John Snowdon

Introduction

Depression in old age has been said to be widespread, common and disa-
bling,1,2 yet the rates of treatment of depression among elderly persons are
markedly lower than among younger adults. 3 Underdiagnosis and under-
treatment of depression in old age have been attributed to a belief, shared
by doctors and patients, that the depressions experienced by elderly persons
are usually a “normal” consequence of the many physical illnesses and social
and economic problems that they endure. Both clinicians and patients may
incorrectly attribute depressive symptoms to the ageing process.3
Blazer4 concluded that the prevalence of clinically significant depressive
symptoms in community samples of older adults is approximately 8% to
15%, while the prevalence of current major depression in such samples is only
1%. He commented5 that divergence between results of studies in the last two
decades concerning the prevalence of late life depression largely derives from
variations in the definitions of “caseness” of depression.
There has been controversy regarding the prevalence of depression in dif-
ferent age groups. Is it higher in young, middle-aged or older persons? Blazer5
suggested that the debate is often meaningless because the varying findings
are based on differing understanding of what is meant by “depression”.
Some used categorical definitions, some recorded whether subjects “scored”
as depressed on dimensional scales, and others developed age-appropriate
284 THE AGEING BRAIN

schedules for eliciting evidence of depressive features. Blazer4 stated that

“although DSM-III-R is not “age biased”, it fails to accommodate correlates
of age such as comorbid cognitive impairment and comorbid physical illness”.
Caine et al.6 commented that affective disturbances often are not expressed
symptomatically among elderly patients in the same stereotypic fashion that
is encountered among younger patient populations, and “current use of rigor-
ous, stereotypic criteria for establishing psychiatric diagnoses may prove to
be highly reliable but not especially valid”.
It could be meaningful to discuss differences between age groups in
the prevalence of a particular type of depression if there is good reason to
believe that a treatment is effective for that type and not for others. It might
be appropriate to investigate whether particular factors (for example, brain
changes) are associated with that type of depression and not with others, and
to consider whether those factors are age-associated. For example, research
on reliably defined psychotic depression or melancholia could provide useful
guides to prevention, treatment and prognosis. Such studies have been con-
ducted.7–10
Contrastingly, reports of differences in prevalence between age groups
that confine consideration to cases that fulfil criteria for major depression,
without considering the clinical context (aetiology, comorbidity, function and
effects), could be considered relatively meaningless. It is possible that older
people fulfilling four rather than five relevant clinical criteria typically have
more serious depressions than younger subjects fulfilling five criteria. We do
not know. Having “major” depression does not necessarily mean having a
more severe or disabling depression. What matters far more to the clinician is
whether the patient has a distressing, disabling condition that could respond
to treatment.
Various experts have emphasised their conclusion that depression (not just
major depression) is less common in old age than among younger age groups.
One big danger in airing such views is that they will lead to reduced incentives
for doctors to be on the lookout for depression among their older patients.
Especially this may apply to patients with comorbid physical illnesses.
Another danger is that administrators may be more inclined to fund services
aimed at preventing, treating and educating about depression in non-elderly
rather than in elderly age groups. The debate may often be meaningless5 but
there is good reason to re-examine the evidence, to ensure that doctors and
administrators are properly informed on what the prevalence studies really
have demonstrated.

Cross-Age Studies

Studies that allow comparisons of rates of depression in different age groups

have led to inconsistent findings. Jorm11 referred to 14 reports of age-asso-
ciated variations in scores on depressive symptom scales. Five showed an
AGE VARIATION IN THE PREVALENCE OF DEPRESSION 285

increase with age, four showed no age-group differences, two reported a fall
followed by a rise, and three showed a decrease. A further study12 (n=2622),
showed a negative correlation between age and scores on two depression
scales, even though certain items (e.g. “Feeling so miserable this interfered
with sleep”, “Feelings of not caring if never woke up”, “Hopelessness”, “Loss
of interest”, “Suicidal thoughts”) were endorsed more often by older people.
The authors suggested that the depression picture for elderly persons may be
characterised by a diminished or reduced evaluation of the future, and that
the nature of the depression experienced by younger and older people may
be qualitatively different.

Table 1. Differences between studies in the age-group recorded as showing peak

prevalence rates of depressive disorders.

Study Date of Country Age-group recording peak

Publication prevalence rate

Regier et al.13 1993 U.S. 6.4% MDE + dysthymia

at 25–44 years

Murphy et al.24 2000 U.S. 3.7% at <45,

1.9% 45–64,
0.9% at 65+

Australian Bureau 1998 Australia Men 35–44,

of Statistics19 Women 18–24

Lindeman et al.21 2000 Finland 11.8% one-year MDE at 45–54,

men 9.3%, women 13.6%
(7.6% at 55–64, 6.7% 65–75)

Bland et al.14 1988 Canada 55–64

Fichter et al.15 1996 Germany 55–64

Lehtinen et al.16 1990 Finland 60–69

Sandanger et al.22 1999 Norway 60–79

Hwu et al.18 1996 Taiwan 1.0% at 65+

(0.7% at 30–65)

Jorm’s review of Various (1)



seven other studies11  (2) no significant age difference
(3)
(4) males no age difference,
female peak 45–64
(5) 30–69 steady, then decrease
(6) 45–64
(7) male 60–69,
female 45–64
286 THE AGEING BRAIN

Jorm11 highlighted inconsistency, too, in the results of research examining

the prevalence of depressive disorders (Table 1). Several groups reported no
age difference. The Epidemiologic Catchment Area (ECA) study13 showed a
peak depression rate (6.4%) at age 25-44 years, though only cases of major
depressive episode and dysthymia were included. Two studies14,15 reported
peak rates at age 55–64 years, and another16 showed a peak at 60–69 years.
The latter study showed a progressive rise in General Health Questionnaire17
scores beyond 65 years, even though there was a reduced prevalence of neu-
rotic depression. A Taiwan study18 reported a prevalence of 0.7% at 30–65
years, and 1.0% at over 65 years. An Australian study19 of 10,000 adults,
which used the Composite International Diagnostic Interview20 (CIDI) and
ICD-10, reported the highest rate of affective disorders among men was at
35–44 years, but among women it was at 18–24 years.
Since Jorm’s review, research from Finland21, using a short form of the
CIDI, has shown 1-year prevalence rates of major depressive episode to be
highest at age 45–54 years in both males (9.3%) and females (13.6%), the
rate for the total sample in the 65–75 years age-group being considerably
lower (6.7%). However, using the CIDI in Norway, the 2-week prevalence
of ICD-10 depression was found to be higher at age 60–79 years (4.2%) than
at age 20-39 (2.1%) or 40–59 (2.5%).22.The prevalence was higher among
women in all three age groups (7.9% versus 1.2% at 60–79 years).
Romanovski et al.23 provided an important additional analysis from one
of the ECA sites. In Eastern Baltimore (n=3481, 923 being over 65 years),
1.2% were diagnosed as having DSM-III major depression at 25–44 years,
2.0% at 45–64 years and 0.5% at over 65 years. In contrast, 3.7% were diag-
nosed as having DSM-III depressive disorders other than major depression
at 25–44 years, with a lower percentage (2.9%) at 45–64 years, and a rise to
5.0% at over 65 years. These authors23 concluded that the total prevalence
of depression increased with age (p<.005) in each sex, though the rates in all
age groups were higher in females.
Jorm11 did not include results from the Stirling County Study24 in his
review. Murphy et al.24 in 1992 used the Diagnostic Interview Schedule25
(DIS), also used in the ECA study, and reported an age-associated decrease
in the one-month prevalence of major depressive episode (MDE) from 3.7%
(<45 years), through 1.9% (45–64) to 0.9% (65+). The prevalence of MDE
plus dysthymia showed, as in the ECA study, a decrease with age in both
sexes. When in 1952, 1970 and 1992 they used criteria other than DSM to
diagnose depression, they found an age-associated increase in the prevalence
among males, but a decrease among females.
Finally, while considering studies that compared age groups, those that
showed differences between middle-aged, “young old” and late life groups
should receive attention (Table 2). In Eastern Baltimore 26 there was an
increase in the major depression rate from 0.7% at 65–74 to 1.3% at over
75 years. Kay et al.27 reported 6.3% at 70–79 years and 15.5% at over 80
years. Beekman et al.28 reported an increase in prevalence of major depres-
AGE VARIATION IN THE PREVALENCE OF DEPRESSION 287

sion from 1.3% at 55-59 years to 2.7% at 80-84 years, and a corresponding
increase in minor depression from 9.4% to 16.7%. Roberts et al.29 showed
an increase in the prevalence of major depressive episode from 8.1% at age
50–59 years and 6.9% at 60–69 years, to 10.4% at 70–79 and 12.7% at 80
years. By contrast, Prince et al,30 analysing data from subjects aged over 65
years in 14 centres in Europe, showed no overall tendency for the prevalence
of depressive ‘caseness’ to rise or fall with age.
Palsson et al.31 used the Comprehensive Psychopathological Rating Scale32
in a longitudinal study of people initially aged 70 years and residing in their
own homes. The one-month prevalence of DSM-III-R depressive disorder

Table 2. Studies showing differences in prevalence of depressive disorders between

“young old” and older samples.

Study Year of Country Type of Prevalence of

publication depressive depressive disorder in
disorder different age-groups

Kramer et al.26 1985 U.S. Major 0.7% 65–74,

1.3% 75+

Kay et al.27 1985 Australia Major 6.3% 70–79,

15.5% 80+

Beekman et al.28 1995 Netherlands Major 1.8% 65–69,

2.7% 80–84

Beekman et al.28 1995 Netherlands Minor 11.4% 65–69,

16.7% 80–84

Roberts et al.29 1997 U.S. Major 6.9% 60–69,

10.4% 70-79,
12.7% 80+

Palsson et al.31 2001 Sweden Depressive Longitudinal study:

disorder 5.6% at 70,
11.2% at 79,
13% at 85

Prince et al.30 1999 Europe “Caseness” No overall tendency to

(14 centres) change with age

65–74 yrs 75–84 yrs 85+yrs

Kennedy et al.57 1989 U.S. CESD 14.1% 18.7% 23.9%

“cases”
Blazer et al.58 1991 U.S. CESD
“cases” 8.1% 10.3% 12.3%
Newman et al.56 1998 Canada GMS- 10% 12.8% 15.3%
AGECAT
288 THE AGEING BRAIN

Table 3. The prevalence in old age of major depression and other depressive disor-
ders (fulfilling DSM criteria): Results from representative studies.

Study Date Place Instrument Age No. of Major Other DSM

subjects depression depressive
(%) disorders
(%)

Blazer & 1980 North OARS 65+ 997 3.7 11

Williams34 Carolina Depression
scale

Kay et al.27 1985 Hobart GMS 70–79 274 6.3 16.5


DSM-III 80+ 15.5 22.4

Kramer  1985 United DIS 65-74 3481 0.7 1.0
et al.26

States DSM-III 75+ 1.3 1.1

Madianos 1992 Athens CESD + 65+ 251 1.6 7.9

et al.35 psychiatric
evaluation

Skoog36 1993 Goteborg Psychiatric 85 347 13.0 6.6

interview
DSM-III-R

Hender- 1993 Canberra CIE 70+ 945 0.4 0.6

son DSM-III-R
et al.33

Lobo 1995 Zaragoza GMS 65+ 1080 1.0 3.8

et al.37 Psychiatric
interview

Beekman 1995 Netherlands CESD 55–85 3056 2.0 12.9

et al.28 + DIS

Fichter 1995 Munich GMS 85+ 358 1.4 5.1

et al.38 + HDRS

Pahkala 1995 Ahtari Zung + GP 65+ 1022 2.2 14.3

et al.39 (Finland) interview

Liu 1997 Rural Psychiatric 65+ 1313 6.1 6.9

et al.40 China interview
DSM-III-R

Gallo 1997 Baltimore DIS 50+ 1612 1.7 15.8

et al.41

Forsell 1998 Stockholm DSM-IV 70+ 1101 7.2 3.5

et al.42

Newman 1998 Edmonton GMS + 65+ 1119 0.9 3.6

et al.43 clinician
interviews
AGE VARIATION IN THE PREVALENCE OF DEPRESSION 289

(including major, dysthymia and depression NOS) was found to rise from
5.6% at age 70 years (n=392; men 1.2%, women 8.8%), 5.9% at age 75
years, and 11.2% at age 79 years (n=206), to 13.8% at age 83 years (n =
116; men 5.6%, women 17.5%) and 13.0% at age 85 years. The incidence of
depression increased from 17 to 44 per 1000 person-years between the ages
of 70–79 and 79–85.

Prevalence Studies of Late Life Depression

In addition to those already mentioned,24,26–29 studies of the prevalence of

DSM major depression in community samples of older people have been
reported from various countries. Inconsistency between results has been
remarkable (Table 3). For example, in contrast to a major depression rate of
11% among a sample aged over 70 years in California,29 Henderson et al.33
reported that 0.4% of a same-age sample in Canberra had major depression
and another 0.6% had current dysthymia. Results from a number of studies
that reported rates of major depression and other DSM depressive disorders
are shown in Table 3.26–28,33–43 Although not exhaustive, the list is believed
to be representative. Only one study from each centre is reported. A more
complete list is provided by Beekman et al.,44 who referred to 34 epidemio-
logical studies of old age depression.
Palsson and Skoog45 noted that total prevalence rates of DSM depres-
sive disorders reported in a number of studies ranged from less than 2% to
26.8%. Beekman et al.44 listed 28 studies that reported rates of major plus
minor depression. Twenty recorded rates in the range 9–18%. Four each
recorded rates above and below this range. Palsson and Skoog45 commented
that epidemiological studies of depression among elderly people tended to be
heavily weighted towards people aged 65–75 years, thus over-shadowing an
increased prevalence after age 75 years.
Less inconsistency is evident between the findings of researchers who
reported rates of clinically significant depression in community samples of
older people, using structured interview schedules. Copeland et al.46 referred
to data from nine centres in Europe when reporting a prevalence rate of
12.3%, and various other studies (Table 4)43,46–55 found rates around the
range 10.3%54 to 13.5%.50
Newman et al.,56 using GMS-AGECAT, reported a rise in prevalence
of depression from 10% at 65–74 years to 12.8% at 75–84 and 15.3% at
over 85 years. Two studies57,58 using the CESD also noted increased rates of
depression across the same three age-groups. The percentages reported by
Kennedy et al.57 were 14.1, 18.7 and 23.9, respectively, and by Blazer et al.58
were 8.1, 10.3 and 12.3.
290 THE AGEING BRAIN

Table 4. The prevalence of depressive symptoms and clinically significant depression

identified during structured interviews.

Study Date Place Instrument Age No. of subjects %

Gurland 1983 London and CARE 65+ 841 12.7

et a.47 New York

Copeland 1987 Liverpool GMS and 65+ 1070 11.3

et al.48 AGECAT

Ben-Arie 1987 Capetown PSE and 65+ 139 13.0

et al.49 CATEGO

Lindesay 1989 London Short-CARE 65+ 890 13.5

et al.50

Livingston 1990 London Short-CARE 65+ 779 17.3

et al.51

Van Ojen 1995 Amsterdam AGECAT 65–84 4051 11.7

et al.52

Kua et al.53 1996 Singapore GMS and 65+ 1062 6.0

AGECAT

Kirby 1997 Dublin GMS and 65+ 1232 10.3

et al.54 AGECAT

Newman 1998 Edmonton GMS and 65+ 1119 11.4

et al.43 AGECAT

Copeland 1999 Nine centres GMS and 65+ 13,808 12.3

et al.46 in Europe AGECAT

Chong 2001 Taiwan GMS and 65+ 1350 21.3

et al.55 AGECAT

Conclusions from these Prevalence Studies

A more than tenfold difference between centres in the prevalence of depres-

sive disorders in old age (Table 3) seems unlikely, though it should be noted
that reported suicide rates of some countries differ by more than tenfold.59
Prevalence rates in the non-elderly populations of different countries also vary
to an astonishing extent.22,60 Possible reasons include the fact that research-
ers differ in their use of diagnostic criteria, and in the methodology used to
establish those criteria.
Palsson and Skoog’s45 comment about differences between samples in the
percentages of “young old” and “old old” included in a study is pertinent.
Flaws in methodology (as discussed61 in relation to the large Australian sur-
AGE VARIATION IN THE PREVALENCE OF DEPRESSION 291

vey19) may affect results, as may differences in response rate between young
and old samples.
Older persons may respond differently from younger people to being ques-
tioned about symptoms and feelings. Older people with depressive disorders
are less likely to acknowledge being sad, down or depressed in mood.62 They
are less likely to admit to feelings of hopelessness or anhedonia.63 Difficulties
with hearing or understanding questions may be more frequent in old age, but
it could be that elderly people are more inclined to misrepresent their feelings
in order to fend off perceived threats to their self-esteem. Studies that rely on
subject-reported symptoms may underestimate the prevalence of disorders
that fulfil DSM criteria.
Diagnostic interview schedules used in cross-age studies have been those
primarily developed for assessing mental disorders among physically well,
young or middle-aged adults. They were not designed to facilitate recogni-
tion of depression precipitated by or associated with physical, cognitive and
environmental changes that become more common in late life. Jorm11 noted
that the DIS and CIDI may discount symptoms that may be attributable to
physical illness. Unless interviewers are enabled to use clinical judgement
(which is not feasible if they are lay interviewers) somatic symptoms may be
mistakenly attributed to physical disorders rather than depression. It is desir-
able to give consideration to the clinical context of all symptoms that could
be depressive even if feelings of depression are denied, and to seek additional
information from those who know the subjects well.
A further source of bias, when examining variation with age in the
prevalence of depression, is that older persons may be more likely to decline
involvement in surveys. Most cross-age studies have not reported whether
response rates differed between age groups. The recent large Australian
study19 did not report differential response rates. Kramer et al.26 noted in
their study that interviews were completed in only 60% of those aged over
65 years, but from over 80% of the non-elderly sample. Henderson et al.64
recorded a response rate of 54% among people aged 60–79 but 70% among
those aged 18-–59 years. Response tendencies can be influenced by cogni-
tive status and the presence of chronic disease65 and disabilities (including
deafness). There is certainly evidence that persons with age-associated brain
disease may be excluded from such surveys: the Australian study19 excluded
persons with moderate or severe dementia and those living in residential
care.
Although response bias may partly account for differences between young
and old in reported prevalence rates of depression, it is also possible that
older people with depression do not fulfil DSM criteria for major depression
so frequently as younger people. This maybe should be the crux of this discus-
sion. Are the data on prevalence of major depression clinically meaningful? Is
it not more meaningful to ascertain the prevalence of “cases” of depression
for whom clinical interventions may be beneficial, whether or not they fulfil
DSM criteria for major depression?
292 THE AGEING BRAIN

Some experts66 have suggested that “depression” means “major depres-

sion”. Various studies of depression prevalence have reported rates only of
subjects with major depression and dysthymia. It is true that people with at
least five of the listed DSM-IV criterion symptoms of depression are likely
to have more severe functional and mental problems than those with only
2–4 such symptoms. However, there is good evidence that many patients
with ‘subsyndromal’ depressive symptoms (i.e. who do not report symptoms
severe or persistent enough to fulfil DSM-IV criteria for major depression
or dysthymia) are functionally disabled to a degree comparable to patients
with major depression or dysthymia.67 Such patients experience psychosocial
dysfunction which improves when they are treated with an antidepressant.68
Their quality of life is worse and their dysfunction and disability are greater
than is true of hypertension or diabetes.69 Although the morbidity associated
with subsyndromal depression may be less than that attributable to major
depression, the prevalence rates are greater and the total attributable burden
to the community is larger for the “subclinical” depressions.70 The number of
days lost from work was similar for people with minor and those with major
depression.71
Beekman et al.72 found the incidence and course of minor or “subsyndro-
mal” depression were closely related to impaired physical health, whereas
major depression was more frequently associated with long-standing vulner-
ability factors, such as family and personal histories of depression. If this is
so, it would suggest that different management strategies are likely to be used,
but it does not mean that one is more treatable than the other.
Having surveyed over 4000 community-dwelling elderly persons, Hybels
et al.73 concluded that depression appears to exist along a continuum. Slater
and Katz70 recommended that clinical care and public policy be designed to
address the needs of those with the wider spectrum of disorders, rather than
concentrating on those with major depression. Gurland et al.47 reported that
13% of elderly people have “pervasive depression” to a degree warranting
clinical intervention. Copeland74 regarded it as inappropriate to use the term
“subsyndromal” when referring to depressions that could be relieved by clini-
cal interventions.
Looking at depression in its broader sense, it was found that handicap due
to disability is associated with a sixfold higher prevalence rate of depression.75
Depression is common among residents of aged care facilities76 and hospital
inpatients77, and in association with dementia,78 strokes,79 and Parkinson’s
disease (PD).80 There is evidence of associations between ageing-related (pos-
sibly vascular) changes in the brain and development of depression.81
Given these findings, and providing the definition of depression is not
restricted to DSM “major depression”, it does not make sense to suggest
that depression is much less common in old age. Most of the studies listed in
Tables 3 and 4, together with cross-age studies that examined rates of clini-
cally significant depression rather than of just major depression, point to a
high prevalence of depression in late life.
AGE VARIATION IN THE PREVALENCE OF DEPRESSION 293

As well as losses of health and physical ability in old age, it has been sug-
gested that late life is commonly a time of loss in other ways. Certainly, loss
situations (such as bereavements, loss of job or change of accommodation)
are relatively common, and loss of role or loss of self-esteem may have par-
ticular relevance in old age. The losses and stressors are largely different from
those of young people,82 among whom relationship problems are common
precipitants of depressive features. It cannot be stated with certainty that
older people experience negative life-events more often than the young, but
providing that ongoing loss or impairment of physical well-being and func-
tion are taken into consideration, surely older people are at least as likely as
younger persons to experience depression-inducing loss situations. This does
not mean they are at least as likely to develop depression. Indeed, some might
argue that “survivors” to old age develop increased resilience as they age.

Summary

The main conclusion to be derived from this discussion is that clinically

significant depression is common in old age, and it would be inappropriate
to use data concerning the prevalence of only major depression and dys-
thymia when planning clinical services. Studies of age-related differences in
the prevalence of depression commonly have not examined the prevalence
of “subsyndromal” depressions in different age groups, even though they
may be just as distressing and disabling as major depressions. What matters,
when determining allocation of mental health resources for management of
depression, is whether those resources, together with other services available
in the community, can help reduce distress, improve quality of life, and have
a beneficial effect on the functional abilities of depressed people.
Findings from cross-age studies have been inconsistent. Errors in case
ascertainment could explain why some of these studies have found the prev-
alence of major depression to be much lower in old age. It is also possible
that major depression (i.e. the prevalence of cases fulfilling the DSM criteria)
is indeed less common in old age. However, there is a need for recognition
that severe, distressing but treatable late life depression commonly does not
fulfil DSM-IV criteria for major depression or dysthymia. It is inappropriate
to imply to doctors and others that depression is far less prevalent in old age
and therefore that there is less need to look out for it. By pointing to the high
prevalence we may provide incentives to improve rates of recognition and
treatment of depression in late life.

References
1. Katona CLE. Depression in old age. Chichester: Wiley, 1994.
2. Lebowitz BD. Depression in late life: directions for intervention research. In: Maj
M, Sartorius N, editors. Depressive disorders. WPA series. Evidence and experi-
ence in psychiatry, Vol. 1. Chichester: Wiley, 1999; 382–384.
294 THE AGEING BRAIN

3. Friedhoff AJ. Consensus development conference statement: diagnosis and treat-

ment of depression in late life. In: Schneider LS, Reynolds CF, Lebowitz BD,
Friedhoff AJ, editors. Diagnosis and treatment of depression in late life. Wash-
ington DC:American Psychiatric Press, 1994; 493–511.
4. Blazer DG. Epidemiology of late-life depression. In: Schneider LS, Reynolds CF,
Lebowitz BD, Friedhoff AJ, editors. Diagnosis and treatment of depression in late
life. Washington DC: American Psychiatric Press, 1994; 9–19.
5. Blazer D. Filling in the gaps about depression in the elderly. In: Maj M, Sartorius
N, editors. Depressive disorders. WPA series. Evidence and experience in psychia-
try. Chichester: Wiley, 1999; 376–377.
6. Caine ED, Lyness JM, King DA, Connors L. Clinical and etiological heterogeneity
of mood disorders in elderly patients. In: Schneider LS, Reynolds CF, Lebowitz
BD, Friedhoff AJ, editors. Diagnosis and treatment of depression in late life.
Washington DC: American Psychiatric Press, 1994; 21–53.
7. Nelson JC, Conwell Y, Kim K, Mazure C. Age at onset in late-life delusional
depression. Am J Psychiat. 1989; 146; 785–786.
8. Nelson JC, Mazure CM, Jatlow PI. Does melancholia predict response in major
depression? J Affect Disorders. 1990; 18: 157–165.
9. Brodaty H, Luscombe G, Parker G, Wilhelm K, Hickie I, Austin MP, Mitchell
P. Increased rate of psychosis and psychomotor change in depression with age.
Psychol Med. 1997; 27:1205–1213.
10. Parker G, Roy K, Hadzi-Pavlovic D, Wilhelm K, Mitchell P. The differential
impact of age on the phenomenology of melancholia. Psychol Med. 2001; 31:
1231–1236.
11. Jorm AF. Does old age reduce the risk of anxiety and depression? A review of epi-
demiological studies across the adult life span. Psychol Med. 2000; 30:11–22.
12. Christensen H, Jorm AF, Mackinnon AJ, Korten AE, Jacomb PA, Henderson
AS, Rodgers B. Age differences in depression and anxiety symptoms: a structural
equation modelling analysis of data from a general population sample. Psychol
Med. 1999; 29:325–339.
13. Regier DA, Farmer ME, Rae DS, Myers JK, Kramer M, Robins LN, George LK,
Karno M, Locke BZ. One-month prevalence of mental disorders in the United
States and sociodemographic characteristics: the Epidemiologic Catchment Area
study. Acta Psychiat Scand. 1993; 88:35–47.
14. Bland RC, Newman SC, Orn H. Period prevalence of psychiatric disorders in
Edmonton. Acta Psychiat Scand. 1988; 78:33–42.
15. Fichter MM, Narrow WE, Roper MT, Rehm J, Elton M, Rae DS, Locke BZ,
Regier DA. Prevalence of mental illness in Germany and the United States. Com-
parison of the Upper Bavarian study and the Epidemiologic Catchment Area
program. J Nerv Ment Dis. 1996; 184:598–606.
16. Lehtinen V, Joukamaa M, Lahtela K, Raitasalo R, Jyrkinen E, Maatela J, Aromaa
A. Prevalence of mental disorders among adults in Finland: basic results from the
Mini Finland Health Survey. Acta Psychiat Scand. 1990; 81:418–425.
17. Goldberg DP. The detection of psychiatric illness by questionnaire. Maudsley
Monograph 21. London: Oxford University Press, 1972.
18. Hwu H-G, Chang I-H, Yeh E-K, Chang C-J, Yeh L-L. Major depressive disorder
in Taiwan defined by the Chinese diagnostic interview schedule. J Nerv Ment Dis.
1996; 184:497–502.
19. Australian Bureau of Statistics. Mental health and wellbeing: Profile of adults,
Australia. 1997. Canberra: Australian Bureau of Statistics, 1998.
20. Robins LN, Wing J, Wittchen HU, Helzer JE, Babor TF, Burke J, Farmer A,
Jablenski A, Pickens R, Regier DA. The Composite International Diagnostic
Interview. Arch Gen Psychiat. 1988; 45:1069–1077.
AGE VARIATION IN THE PREVALENCE OF DEPRESSION 295

21. Lindeman S, Hämäläinen J, Isometsä E, Kapiro J, Poikolainen K, Heikkinen M,

Aro H. The 12-month prevalence and risk factors for major depressive episode
in Finland: representative sample of 5993 adults. An epidemiologic instrument
suitable for use in conjunction with different diagnostic systems and in different
cultures. Acta Psychiat Scand. 2000; 102:178–184.
22. Sandanger I, Nygård JF, Ingebrigtsen G, Sørensen T, Dalgard OS. Prevalence,
incidence and age at onset of psychiatric disorders in Norway. Soc Psych Psych
Epid. 1999; 34:570–579.
23. Romanovski AJ, Folstein MF, Nestadt G, Chahal R, Merchant A, Brown CH,
Gruenberg EM, McHugh PR. The epidemiology of psychiatrist-ascertained
depression and DSM-III depressive disorders. Results from the Eastern Baltimore
Mental Health Survey Clinical Reappraisal. Psychol Med. 1992; 22:629–655.
24. Murphy JM, Laird NM, Monson RR, Sobol AM, Leighton AH. A 40-year
perspective on the prevalence of depression. Arch Gen Psychiat. 2000; 57:
209–215.
25. Robins LN, Helzer JE, Croughan J, Ratcliff KS. National Institute of Mental
Health Diagnostic Interview Schedule: its history, characteristics and validity.
Arch Gen Psychiat. 1981; 38:381–389.
26. Kramer M, German PS, Anthony JC, Von Korff M, Skinner EA. Patterns of men-
tal disorders among the elderly residents of Eastern Baltimore. J Am Geriatric
Soc. 1985; 33:236–245.
27. Kay DWK, Henderson AS, Scott R, Wilson J, Rickwood D, Grayson DA. The
prevalence of dementia and depression among the elderly living in the Hobart
community: the effect of the diagnostic criteria on the prevalence rates. Psychol
Med. 1985; 15:771–778.
28. Beekman ATF, Deeg DJH, van Tilburg T, Smit JH, Hooijer C, van Tilburg W.
Major and minor depression in later life: a study of prevalence and risk factors.
J Affect Disorders. 1995; 36:65–75.
29. Roberts RE, Kaplan GA, Shema SJ, Strawbridge WJ. Does growing old increase
the risk of depression? Am J Psychiat. 1997; 154:1384–1390.
30. Prince MJ, Beekman AT, Deeg DJ, Fuhrer R, Kivela SL, Lawlor BA, Lobo A, Mag-
nusson H, Meller I, van Oyen H, Reischies F, Roelands M, Skoog I, Turrina C,
Copeland JF. Depression symptoms in late life assessed using the EURO-D scale.
Effect of age, gender and marital status in 14 European centres. Br J Psychiat.
1999; 174:339–345.
31. Palsson SP, Östling S, Skoog I. The incidence of first-onset depression in a popula-
tion followed from the age of 70 to 85. Psychol Med. 2001; 31:1159–1168.
32. Asberg M, Perris C, Schalling D, Sedvall G. The CPRS – development and applica-
tions of a psychiatric rating scale. Acta Psychiat Scand. 1978; suppl 271.
33. Henderson AS, Jorm AF, Mackinnon A, Christensen H, Scott LF, Korten AE,
Doyle C. The prevalence of depressive disorders and the distribution of depres-
sive symptoms in later life: a survey using Draft ICD-10 and DSM-III-R. Psychol
Med. 1993; 23:719–729.
34. Blazer D, Williams CD. Epidemiology of dysphoria and depression in an elderly
population. Am J Psychiat. 1980; 137:439–444.
35. Madianos MG, Gournas G, Stefanis CN. Depressive symptoms and depression
among elderly people in Athens. Acta Psychiat Scand. 1992; 86:320–326.
36. Skoog I. The prevalence of psychotic, depressive and anxiety syndromes in
demented and non-demented 85-year-olds. Int J Geriatr Psych. 1993; 8:247–
253.
37. Lobo A, Saz P, Marcos G, Dia J-L, De-la-Cámara C. The prevalence of dementia
and depression in the elderly community in a Southern European population.
Arch Gen Psychiat. 1995; 52:497–506.
296 THE AGEING BRAIN

38. Fichter MM, Bruce ML, Schroppel H, Meller I, Merikangas K. Cognitive impair-
ment and depression in the oldest old in a German and in U.S. communities. Eur
Arch Psychiat Clin N. 1995; 245:319–325.
39. Pahkala K, Kesti E, Köngäs-Saviaro P, Laippala P, Kivelä S-L. Prevalence of
depression in an aged population in Finland. Soc Psych Psych Epid. 1995; 30:
99–106.
40. Liu CY, Wang SJ, Teng EL, Fuh JL, Lin CC, Lin KN, Chen HM, Lin CH, Wang
PN, Yang YY, Larson EB, Chou P, Liu HC. Depressive disorders among older
residents in a Chinese rural community. Psychol Med. 1997; 27:943–949.
41. Gallo JJ, Rabins PV, Lyketsos CG. Depression without sadness: functional outcomes
of nondysphoric depression in later life. J Am Geriatr Soc. 1997; 45:570–578.
42. Forsell Y, Jorm AF, Winblad B. The outcome of depression and dysthymia in a
very elderly population: results from a three-year follow-up study. Aging Ment
Health. 1998; 2:100–104.
43. Newman SC, Sheldon CT, Bland RC. Prevalence of depression in an elderly com-
munity sample: a comparison of GMS-AGECAT and DSM-IV diagnostic criteria.
Psychol Med. 1998; 28:1339–1345.
44. Beekman ATF, Copeland JRM, Prince MJ. Review of community prevalence of
depression in later life. Br J Psychiat. 1999; 174:307–311.
45. Palsson S, Skoog I. The epidemiology of affective disorders in the elderly: a
review. Int Clin Psychopharm. 1997; 12: (Suppl 7): S3–S13.
46. Copeland JR, Beekman AT, Dewey ME, Hooijer C, Jordan A, Lawlor BA, Lobo
A, Magnusson H, Mann AH, Meller I, Prince MJ, Reischies F, Turrina C, de Vries
MW, Wilson KC. Depression in Europe. Geographical distribution among older
people. Br J Psychiat. 1999; 174:312–321.
47. Gurland B, Copeland J, Kuriansky, J, Kelleher M, Sharpe L, Dean LL. The mind
and mood of aging. London: Croom Helm, 1983.
48. Copeland JRM, Dewey ME, Wood N, Searle R, Davidson IA, McWilliam C.
Range of mental illness among the elderly in the community. Prevalence in Liv-
erpool using the GMS-AGECAT package. Br J Psychiat. 1987; 150:169–174.
49. Ben-Arie O, Swartz L, Dickman BJ. Depression in the elderly living in the com-
munity: its presentation and features. Br J Psychiat. 1987; 150:169–174.
50. Lindesay J, Briggs K, Murphy E. The Guy’s / Age Concern Survey. Prevalence
rates of cognitive impairment, depression and anxiety in an urban elderly com-
munity. Br J Psychiat. 1989; 155:317–329.
51. Livingston G, Hawkins A, Graham N, Blizard B, Mann A. The Gospel Oak Study:
prevalence rates of dementia, depression and activity limitation among elderly
residents in Inner London. Psychol Med. 1990; 20:137–146.
52. Van Ojen R, Hooijer C, Jonker C, Lindeboom J, van Tilburg W. Late-life depres-
sive disorder in the community, early onset and the decrease of vulnerability with
increasing age. J Affect Disorders. 1995; 33:159–166.
53. Kua EH, Ko SM, Fones C, Tan SL. Comorbidity of depression in the elderly – an
epidemiological study in a Chinese community. Int J Geriatr Psychiat. 1996; 11:
699–704.
54. Kirby M, Bruce I, Radic A, Coakley D, Lawlor BA. Mental disorders among the
community-dwelling elderly in Dublin. Br J Psychiat. 1997; 171:369–372.
55. Chong MY, Chen CS, Tsang HY, Chen CC, Yeh TL, Lee YH, Lo HY. Community
study of depression in old age in Taiwan. Br J Psychiat. 2001; 178:29–35.
56. Newman SC, Bland RC, Orn HT. The prevalence of mental disorders in the
elderly in Edmonton: a community survey using GMS-AGECAT. Can J Psychiat.
1998; 43:910–914.
57. Kennedy GJ, Kelman HR, Thomas C, Wisniewski W, Metz H, Bijur PE. Hierar-
chy of characteristics associated with depressive symptoms in an urban elderly
sample. Am J Psychiat. 1989; 146:220–225.
AGE VARIATION IN THE PREVALENCE OF DEPRESSION 297

58. Blazer D, Burchett B, Service C, George LK. The association of age and depres-
sion among the elderly: an epidemiologic exploration. J Gerontol. 1991; 46:
M210–215.
59. Mo,cicki EK. Epidemiology of suicide. Int Psychogeriatrics 1995; 7: 137-148.
60. Ayuso-Mateos JL, Vázquez-Barquero JL, Dowrick C, Lehtinen V, Dalgard OS,
Casey P, Wilkinson C, Lasa L, Page H, Dunn G, Wilkinson G, ODIN Group.
Depressive disorders in Europe: prevalence figures from the ODIN study. Br J
Psychiat. 2001; 179:308–316.
61. Snowdon J, Draper B, Chiu E, Ames D, Brodaty H. Surveys of mental health and
wellbeing: critical comments. Australas Psychiat. 1998; 6:246–247.
62. Henderson AS. Does ageing protect against depression? Soc Psych Psych Epid.
1994; 29:107–109.
63. Lyness JM, Cox C, Curry J, Conwell Y, King DA, Caine ED. Older age and the
underreporting of depressive symptoms. J Am Geriatr Soc. 1995; 43:216–22.
64. Henderson AS, Jorm AF, Korten AE, Jacomb P, Christensen H, Rodgers B.
Symptoms of depression and anxiety during adult life: evidence for a decline in
prevalence with age. Psychol Med. 1998; 28:1321–1328.
65. Norton MC, Breitner JCS, Welsh KA, Wyse BW. Characteristics of nonresponders
in a community survey of the elderly. J Am Geriatr Soc. 1994; 42:1252–1256.
66. Reifler BV. Depression: diagnosis and comorbidity. In: Schneider LS, Reynolds
CF, Lebowitz BD, Friedhoff AJ, editors. Diagnosis and treatment of depression
in late life. Washington DC: American Psychiatric Press, 1994; 55–59.
67. Lyness JM, King DA, Cox C, Yoediono Z, Caine ED. The importance of sub-
syndromal depression in older primary care patients: prevalence and associated
functional disability. J Am Geriatr Soc. 1999; 47:647–652.
68. Rapaport MH, Judd LL. Minor depressive disorder and sybsyndromal depressive
symptoms: functional impairment and response to treatment. J Affect Disorders.
1998; 48:227–232.
69. Wells KB, Stewart A, Hays RD, Burnam MA, Rogers W, Daniels M, Berry S,
Greenfield S, Ware J. The functioning and well-being of depressed patients.
Results from the Medical Outcomes Study. J Amer Med Assoc. 1989; 262:
914–919.
70. Slater SL, Katz IR. Prevalence of depression in the aged: formal calculations ver-
sus clinical facts. J Am Geriatr Soc. 1995; 43:78–79.
71. Broadhead WE, Blazer DG, George LK, Tse CK. Depression, disability days and
days lost from work in a prospective epidemiologic survey. J Amer Med Assoc.
1990; 264:2524–2528.
72. Beekman AFT, Penninx BWJH, Deeg DJH, Ormel J, Braam AW, van Tilburg W.
Depression and physical health in later life: results from the Longitudinal Aging
Study Amsterdam (LASA). J Affect Disorders. 1997; 46:219–231.
73. Hybels CF, Blazer DG, Pieper CF. Toward a threshold for subthreshold depres-
sion: an analysis of correlates of depression by severity of symptoms using data
from an elderly community sample. Gerontologist. 2001; 41:357–365.
74. Copeland JRM. Depression in old age: diagnostic problems, new knowledge and
neglected areas. In: Maj M, Sartorius N, editors. Depressive disorders. WPA
Series. Evidence and experience in psychiatry, Vol. 1. Chichester: Wiley, 1999;
363–366.
75. Prince MJ, Harwood RH, Blizard RA, Thomas A, Mann AH. Impairment, dis-
ability and handicap as risk factors for depression in old age. The Gospel Oak
Project V. Psychol Med. 1997; 27:311–321.
76. Ames D. Depression in nursing and residential homes. In: Chiu E, Ames D,
editors. Functional psychiatric disorders of the elderly. Cambridge: Cambridge
University Press, 1994.
298 THE AGEING BRAIN

77. Koenig HG, George LK, Peterson BL, Pieper CF. Depression in medically ill
hospitalized older adults: prevalence, characteristics, and course of symptoms
according to six diagnostic schemes. Am J Psychiat. 1997; 154:1376–1383.
78. Burns A, Jacoby R, Levy R. Psychiatric phenomena in Alzheimer’s disease. III.
Disorders of mood. Br J Psychiat. 1990; 157:81–86.
79. Burvill P. Psychiatric aspects of cerebro-vascular disease. In: Chiu E, Ames D,
editors. Functional psychiatric disorders of the elderly. Cambridge: Cambridge
University Press, 1994.
80. Mayeux R. Depression in the patient with Parkinson’s disease. J Clin Psychiat.
1990; 51: (Suppl 7):20–23.
81. Hickie I, Scott E. Late-onset depression: a preventable cerebrovascular disease?
Psychol Med. 1998; 28:1007–1013.
82. Foster JR. Successful coping, adaptation and resilience in the elderly: an interpre-
tation of epidemiologic data. Psychiatr Quart. 1997; 68:189–219.
Chapter 18

VASCULAR DEMENTIA
Perminder Sachdev

Introduction

The importance of vascular factors in the aetiology of cognitive impair-

ment was recognised over a century ago 1, but there was debate regarding
the mechanisms by which this occurred. Alzheimer 2 described “arterio-
sclerotic cerebral atrophy” as a cause of senility around the same time as
Binswanger’s3 paper on “encephalitis subcorticalis chronica progressiva”
was published in 1884. A close examination of Binswanger’s report suggests
that the diagnosis was not certain. The patient, a man in his mid-fifties with
a history of syphilis, presented with a progressive decline in speech and
memory, as well as depression and personality change. On post-mortem
examination, there was minimal atherosclerosis, enlargement of lateral ven-
tricles, marked atrophy of the cerebral white matter, granular deposits on
basal dura mater, and multiple ependymal thickenings. 3 It was Alzheimer4
who attributed this white matter change to atherosclerosis when he described
an analogous case in 1902. Modern reviewers consider Binswanger’s case to
have been one of neurosyphilis,5,6 but the concept of extensive atheroscle-
rotic white matter disease associated with cognitive decline became known
as Binswanger’s disease.
Alzheimer’s7 description of plaques and tangles as a basis of dementia
in the first decade of this century did not greatly reduce the emphasis on
vascular factors that were still considered to be the more common cause of
dementia. Half a century later, the idea that critical atherosclerotic narrow-
ing of cerebral arteries commonly led to progressive ischaemia, neuronal loss
and dementia still held sway although Alzheimer’s disease (AD) had by now
found its way into most neurological texts. A number of developments in the
last three decades have led to a radical rethinking of the association between
cerebrovascular disease (CVD) and dementia, and set the stage for a recon-
ceptualisation of vascular dementia.
300 THE AGEING BRAIN

First, the classic studies of Tomlinson et al.8 in the 1960s clearly dis-
tinguished Alzheimer’s disease from dementia due to vascular causes, and
demonstrated the importance of cerebral infarction for the latter. They also
reported a relationship between the volume of infarcted cerebral tissue and
the likelihood of clinical dementia. Volume loss of up to 100 ml was fre-
quently found in control subjects without dementia, and those with athero-
sclerotic dementia (n=6) had a mean volume of infarction of 186 ml (range
101–412 ml).
Second, work in the mid-1970s9,10 emphasized that dementia from vascu-
lar causes was due to discrete and multiple lesions, most of which resulted
from thromboembolism originating in the extracranial arteries and the heart.
Dementia due to vascular causes became equated with multi-infarct dementia
(MID) although the term was used infrequently. An Ischaemic Score, 10 which
incorporated risk factors for multiple stroke was proposed to differentiate
MID from AD.
Third, the ageing of the population increased the public health importance
of dementia, and epidemiological studies established vascular causes as the
second most common aetiology of dementia after AD in Western countries
and, in fact, the most common cause in some Asian countries.11
Fourth, recent developments in neuroimaging, in particular computed tom-
ography (CT) and magnetic resonance imaging (MRI), have permitted the
examination of patients’ brains for lesions hitherto possible only at autopsy.
It became obvious that cognitive impairment could be caused by subcortical
basal ganglia and white matter lesions, often detected as hyperintensities on
T2-weighted MRI, and even by strategically placed single infarcts. Not all
patients with white matter lesions (WML) had Binswanger’s disease, and the
term “leukoaraiosis” (LA)12 (literally, thinning of the white matter) was pro-
posed to describe these brain abnormalities on neuroimaging. It thus became
accepted that MID was but one form of the dementias of vascular aetiol-
ogy.
Fifth, the term Vascular Dementia (VaD) emerged as the preferred term for
this group of disorders, and some international efforts were made to stand-
ardise its diagnostic criteria.13–17 These efforts resulted in a reconceptualisa-
tion of the pathogenetic mechanisms to VaD.
Sixth, the potential for modifying the risk factors for VaD had become
apparent, raising the possibility of influencing the incidence and prevalence
of dementia in the elderly population. VaD was hailed as a preventable
dementia, which should have much greater recognition than it has hitherto
received.
Seventh, post-mortem studies established that many patients with dementia
had a mixed pathology. It also emerged that factors traditionally recognised
as those increasing the risk of cerebrovascular disease, such as hypertension,
diabetes mellitus, cholesterol and homocysteine, were also possibly risk fac-
tors for AD.
VASCULAR DEMENTIA 301

The Definition of Vascular Dementia

There are two obvious steps in the diagnosis of VaD — the diagnosis of
dementia and the establishment of its vascular aetiology — both of which are
controversial and lack consensus.
i) Defining dementia: Most modern diagnostic systems define dementia
as a multifaceted decline in cognitive functioning from a previous higher
level which impairs functioning in daily life, but is not necessarily progres-
sive. Impairment of memory is usually considered necessary for dementia
according to these criteria, with rare exceptions. 18 The criteria differ in
whether one, two or more other cognitive domains must have impairments
in order to qualify for a diagnosis of dementia. The two most commonly
used criteria for dementia are from the Diagnostic and Statistical Manual
of Mental Disorders–Fourth Edition (DSM-IV) 17 and the tenth revision of
the International Classification of Diseases (ICD-10). 16 The two systems
broadly concur in their emphasis on a decline from a previously higher
level of cognitive function that impairs daily functioning in occupational
or social dimensions. However, there is a point of difference in that DSM-
IV accepts impairment of memory plus one other area of higher cortical
function, whereas ICD-10 requires impairment of memory and two other
areas of cognitive function. Moreover, these criteria do not operationalise
the definition of impairment, so that deficits that that have little impact on
the functioning of a retiree who does not engage in much social activity may
prove to be devastating for a middle-aged man who runs his own business
or manages a company.
ii) Defining Vascular Dementia: That a particular dementia is primarily
due to vascular factors is not easy to establish, and different approaches have
been proposed. One popular approach to diagnose VaD in the past has been
to use a quantitative rating such as the Ischaemia Scale score. Hachinski et
al.10 were the first to propose such a scale comprising 13 items (maximum
possible score 18) based on history and clinical examination (Table 1), and
reported that a score of 7 or more suggested MID and that of 4 or less AD.
This approach has been validated in its ability to distinguish a relatively
pure MID from AD or mixed dementia.19 However it has limitations in that
it focuses on MID to the exclusion of other subtypes of VaD and uses only
some of the relevant clinical information, disregarding neuroimaging and
other laboratory data. Modifications of the Hachinski Ischaemia Scale have
been presented to include neuroimaging data.13 The specific items of the Scale
were recently examined using logistic regression models.20 The items that best
distinguished MID from AD were: stepwise deterioration (odds ratio = 7.8),
fluctuating course (6.1), hypertension (4.2), atherosclerosis (2.6), and focal
neurological symptoms (5.4). MID was distinguished from mixed dementia
by stepwise deterioration (4.6) alone, and fluctuating course (0.2) and history
of stroke (0.08) distinguished AD from mixed dementia. Nocturnal confu-
sion, preservation of personality, depression, and somatic complaints had no
302 THE AGEING BRAIN

Table 1. Hachinski’s Ischaemia Scale.10

Item Score

Abrupt onset 2
Stepwise deterioration 1
Fluctuating course 2
Nocturnal confusion 1
Relative preservation of personality 1
Depression 1
Somatic complaints 1
Emotional incontinence 1
History of presence of hypertension 1
History of strokes 2
Evidence of associated arteriosclerosis 1
Focal neurological symptoms 2
Focal neurological signs 2

discriminating value in this study. Loeb13 included CT brain scan abnormali-

ties on his modified ischaemic score.
The DSM-IV criteria for VaD require that criteria for dementia be met
and the patient have focal neurological signs and symptoms or laboratory
evidence indicative of cerebrovascular disease (e.g., multiple infarctions
involving cortex and underlying white matter) that are judged to be aetiologi-
cally related to the disturbance. Neuroimaging is therefore not an absolute
requirement for the DSM-IV diagnosis of VaD. The ICD-10 criteria require
unequal distribution of deficits in higher cognitive functions, evidence of
focal brain damage (unilateral spastic weakness of the limbs, unilaterally
increased tendon reflexes, an extensor plantar response, pseudobulbar palsy)
and significant cerebrovascular disease judged to be aetiologically related to
the dementia, and again do not specify neuroimaging requirements.
Of particular interest are two recent proposals to develop research diag-
nostic criteria. The Alzheimer’s Disease Diagnostic and Treatment Centers
(ADDTC) of the State of California criteria for VaD14 are notable for not
specifically requiring memory impairment for the diagnosis of dementia; dis-
turbance in any two or more cognitive domains is sufficient for this purpose.
The AADTC criteria only deal with ischaemic lesions, and require that the
patient have evidence of two or more ischaemic strokes, and in case of a single
stroke, a clearly documented temporal relationship. There is a requirement of
at least one infarct outside the cerebellum on neuroimaging. Non-infarction,
ischaemic white matter lesions are insufficient for a diagnosis of Probable
VaD, but may support Possible VaD.
The currently most commonly used criteria in research are the NINDS-
AIREN criteria15 which were developed by a group of 54 neurologists and
neuroscientists as a starting point for international research and discussion.
Like the ICD-10 approach,16 the definition of dementia requires dysfunction
VASCULAR DEMENTIA 303

Table 2. Summary of the NINDS-AIREN diagnostic criteria for Vascular Dementia

(VaD).15

I. Probable VaD: All of the following:

1. Dementia:
(a) Cognitive decline from higher level, and
(b) Impairment of memory, and
(c) Impairment in 2 or more cognitive domains, and
(d) Impairment interferes with daily living.
Exclude if disturbance of consciousness, delirium, psychosis, severe aphasia,
or sensorimotor impairment preclude neuropathological testing, or other brain
disease exists (e.g. AD).
2. Cerebrovascular disease (CVD)
(a) Focal signs on neurological examination, and
(b) Brain imaging (CT/MRI) evidence — multiple or strategic single infarcts,
multiple lacunes, extensive white matter lesions, or combinations.
3. Relationship between CVD and dementia: one or more of the following:
(a) Onset of dementia within three months of stroke, or
(b) Abrupt deterioration, or
(c) Fluctuating, stepwise progression.

II. Possible VaD:

Criteria I-1 and I-2a above, but I-2b lacking, or I-3a, b or c lacking.

III.Definite VaD:
(a) Clinical criteria of Probable VaD, and
(b) Histopathological evidence of CVD (autopsy or biopsy), and
(c) Absence of neuritic plaques and neurofibrillary tangles exceeding those expected
for age, and
(d) Absence of other pathology possible causing dementia.

IV. AD with CVD (instead of “mixed dementia”)

Patients meeting criteria for AD with clinical or imaging evidence of CVD.

of memory and at least two other cognitive domains. The criteria recognise
that the lesions could be ischaemic of haemorrhagic. The presence of cer-
ebrovascular disease is recognised by the presence of focal signs on neuro-
logical examination consistent with stroke, and evidence on neuroimaging.
The latter could be multiple large-vessel strokes, single strategically placed
stroke, lacunar infarcts, extensive WMLs, or a combination of these. The
establishment of a relationship between CVD and dementia was considered
to be problematic — it required either the onset of dementia within three
months of a stroke, or a history of abrupt deterioration in cognitive function,
or fluctuating, step-wise progression of deficits. The criteria acknowledged
that a diagnosis of Definitive VaD could only be made on histopathological
confirmation of significant CVD in the absence of Alzheimer-type pathology.
Clinically, the diagnosis was either Probable or Possible depending upon the
level of certainty of the CVD and its relationship to dementia. These criteria
also argue against the use of “mixed” dementia as a diagnosis, suggesting
304 THE AGEING BRAIN

instead the appellation of “AD with CVD”, thus shifting the causality of
dementia to AD in such cases, a proposal which has been criticised as being
premature21. The criteria do not include haemorrhagic lesions in the aetiology
of dementia even though the workshop recognised the importance of such
lesions. For the diagnosis of dementia, disturbance of consciousness is exclu-
sionary, but the proposal not to make a diagnosis of dementia in the presence
of psychosis, severe aphasia or severe sensorimotor impairment precluding
neuropsychological testing in the NINDS-AIREN criteria is controversial as
intellectual decline in these patients can often be demonstrated by methods
other than formal neuropsychological testing. A summary of the NINDS-
AIREN criteria is presented in Table 2.
The sensitivity and specificity of the criteria vary considerably. In a fol-
low-up study of 148 stroke patients, Desmond et al.2 reported impairment in
one cognitive domain in 19.7%, two domains in 17.0%, and three or more
domains in 15.7%, underscoring the importance of the operational definition.
How cognitive impairment is quantified remains highly variable, with the
sophistication of the neuropsychological battery being an important deter-
minant of the range of deficits recorded. Obviously, the use of mental status
tests, such as the Mini-metal State Examination (MMSE),23 is not considered
to be adequate. The different criteria do not overlap, and prevalence rates
therefore would differ depending upon which criteria set is used. The DSM-IV
criteria are the least restrictive, the ADDTC more sensitive and the NINDS-
AIREN more specific.24,25

The Concept of Vascular Cognitive Impairment (VCI)

We have already examined some of the limitations of the concept of demen-

tia. In addition to the difficulties in operationalising deficits, deciding which
cognitive domains are critical, and what level of impairment is necessary,
the diagnosis may well be too late in the course of the disorder. If VaD is
to be prevented, its recognition should be at a much earlier stage. Hence
the proposal26 to broaden the concept to Vascular Cognitive Impairment
(VCI) which would include the whole range of individuals from those at
high risk of CVD (brain-at-risk) to those meeting strict criteria for VaD.
The syndrome is defined clinically, based on neuropsychological perform-
ance, and vascular and interaction is documented. It is a heterogeneous
concept with varied aetiology, and its specific criteria need development
and refinement.

Epidemiology
VaD is generally considered to be the second most common cause of demen-
tia, accounting for 10–50% of all dementia cases, depending upon the cri-
teria used and the population examined. Prevalence rates have varied across
studies owing to methodological differences, and range from 1.2 to 4.2% in
VASCULAR DEMENTIA 305

persons aged 65 years and over. In the Canadian Health and Ageing Study,27
the prevalence of VaD in persons over 65 years was 1.5%. Another 0.9%
had mixed VaD and AD, and 2.4% were noted to have cognitive impairment
of vascular origin but did not meet criteria for dementia, thereby giving an
overall prevalence of about 5% for VCI. This compared with a prevalence of
5.1% for AD without a vascular component. Data from the European col-
laborative study28 put the prevalence of VaD in individuals aged 65 years and
over at 1.6%, which accounted for 15.8% of all dementia cases. Overall, the
studies vary considerably in the prevalence rates, from 0.0 to 0.8% at age 65
years to 2% to 8.3% at age 90 years. The prevalence was higher in men up
to the age of 85 years, after which it was higher in women. There is a cross-
national effect with AD being more common in Western countries, and VaD
being much more common in Japan, China and Russia. Hagnell et al.,29 in
a Swedish study, estimated the lifetime risk of VaD as 34.5% for men and
19.4% for women. Autopsy data from Western countries confirm VaD to be
the second commonest form of dementia after AD.8,30
In community based studies, the incidence of MID29,31 or “arterioscle-
rotic psychosis”32 has ranged from 0.17 to 0.71/ 100 person-years. In a
hospitalised ischaemic stroke sample, 33 the incidence of VaD was 8.4/100
person-years. The incidence increases with age, and in the European col-
laborative study,26 the rates were 0.5%, 0.8%, 1.9%, 2.4%, 2.4% and
3.0% in the various five-year age bands between 65 and 69 years and 90+
years.

Dementia after stroke

Since there is a close relationship between stroke and VaD, many studies
have examined the rates of dementia in stroke cohorts. The rates at three
months after stroke in two studies were 26.3% (New York study)34 and
25% (Helsinki study).35 Rates from our Sydney Stroke Study were compa-
rable (24%), with another 36% having VCI that did not meet criteria for
dementia (unpublished data). Stroke may thus be considered to be the most
important risk factor for VaD. In the New York study, the rates of VaD in
stroke patients over 60 years were nine times that in controls, with an odds
ratio of 31.2 in the 70–79 years age group. Of all the cases of dementia in this
group, stroke was considered to be the underlying cause in 56.1% cases, and
in 36.4% there was considered to be a combined effect of stroke and Alzhe-
imer’s disease. The incidence of new-onset dementia also increases after an
index stroke. In the New York study, the risk of new-onset dementia one year
after stroke was 5.4% in patients over 60 years, and 10.9% in those over 90
years. In a community-based study from Rochester, Minnesota (USA),36 the
rate for new-onset dementia in the first year after a stroke was 8.6%. In this
study, the incidence of AD also doubled after a stroke, raising issues about
the interaction of the two pathologies. The risk if dementia is also increased
after lacunar stroke, which was 23.1% in the four years after such infarction
in one study,37 a 4-–12 fold increase over controls.
306 THE AGEING BRAIN

Risk factors for Vascular Dementia

Factors related to increased risk of stroke
Since stroke is one of the major determinants of VaD, it is reasonable to
expect that risk factors for stroke would also increase the risk of VaD. It has
been demonstrated38,39 that hypertension, diabetes mellitus, coronary artery
disease, atrial fibrillation, cigarette smoking and dyslipidemia are the major
risk factors for VaD. The role of other factors such as various cardiac fac-
tors, obesity, fibrinogen, antiphospholipid antibodies, etc. remain to be fully
investigated. Sociodemographic factors relevant to stroke risk are age, male
sex and race/ethnicity (e.g., Asians, Afro-Americans).
Hypertension is the most important modifiable risk factor for stroke
and VaD, as it accounts for about one-quarter to one-half of all strokes.40
Hypertension in early life has been associated with the cognitive impairment
and dementia later in life in a number of studies. In the Honolulu-Asia Aging
Study,41 systolic hypertension in mid-life increased the risk of cognitive
deficits later in life. For every 10 mm Hg increase in systolic blood pressure,
there was an increase in risk of 7% for intermediate cognitive function and
5% for poor cognitive function. The association between hypertension and
poor cognitive function was also found in a number of other studies, both
cross-sectional and longitudinal.42–44 There is also some evidence that treat-
ment of hypertension has a protective effect on cognition.45 The association
appears to be not simply for VaD but also for AD.42 Moreover, lower blood
pressure (<130 mm Hg systolic and <75 mm Hg diastolic) may be associated
with cognitive impairment in the elderly.46 While hypertension increases the
risk of VaD, once dementia is set in, high systolic blood pressure may serve
a protective role.39 There may therefore be a J-shaped association between
hypertension and cognition.47 There are many mechanisms by which hyper-
tension may cause cognitive impairment. There is an increase in atheroscle-
rotic disease, leading to more strokes. There is also an increase in arterio-
sclerotic disease, which is a major factor in the development of white matter
pathology.48

Factors related to increased white matter lesions (WMLs)

The main risk factor for WMLs is recognised to be hypertension, presumably
through lipohyalinosis and thickening of the walls of the small perforating
arterioles, thereby causing ischaemic damage to the myelin sheath of axons.49
However, hypoperfusion and hypoxic-ischaemic episodes can also produce
similar lesions. Other risk factors for WMLs include general vascular disor-
ders, diabetes mellitus, and other risk factors for atherosclerosis, high plasma
viscosity and indications of blood-brain barrier damage.

Genetic factors
Genetic factors for CVD, and consequently VaD, are not well understood.
Exceptions are rare disorders such as cerebral autosomal dominant arterio-
pathy with subcortical infarct and leukoencephalopathy (CADASIL)50 and
VASCULAR DEMENTIA 307

autosomal dominant hereditary cerebral haemorrhage with amyloidosis

— Dutch type.51 CADASIL is an arteriopathy due to Notch3 gene mutations
on chromosome 19 leading to confluent WMLs and multiple strokes by the
age of 40–50 years.50 Other rare hereditary vascular conditions unlinked to
Notch3 gene mutation have been identified,52 as have non-CADASIL Bin-
swanger-like syndromes without arterial hypertension.53 Several families with
autosomal recessive leukoencephalopathy and hereditary cerebral amyloid
angiopathy have been described. The role of apolipoprotein E polymorphism
in VaD is unclear, with published studies producing conflicting evidence for
a link with ε4 allele.3,54

Other factors
Stroke is not the only mechanism for the development of VaD, and many
stroke patients do not go on to develop dementia, suggesting that the nature
and extent of stroke and some host factors may be important. Age has consist-
ently been reported as a substantial risk factor for VaD,55 and VaD is more
common in men.29 Certain race/ethnic factors seem to be relevant, with the
risk for VaD being higher in Japan, China and Russia when compared with
Western societies.55 As for AD, education has been suggested as a protective
factor against the development of VaD.33,38,39 It is not known whether this is
because of a threshold effect in the more educated, or education in fact has a
positive effect on brain development, and whether early education or a more
complex and challenging occupation is the key variable.

Clinical-Pathological Correlates and Pathogenesis

Brain parenchymal lesions may be produced through ischaemia, haemorrhage

or oedema because of the involvement of small and/or large arteries resulting
in cortical and/or subcortical infarcts or non-infarct lesions. The major vascu-
lar mechanisms involved are listed in Table 3, and the resulting neuropathol-
ogy will vary according to the dominant mechanisms. Since the classic studies
of Tomlinson et al.,8 the major focus has been on cerebral infarcts due to the
occlusion of large or medium-sized arteries. The earlier suggestion was that
the volume of cerebral tissue infarcted was critical, with dementia most likely
to occur when a threshold of 100 ml was reached and the brain’s compensa-
tory capabilities overwhelmed. Other authors came to a similar conclusion,
although infarct volumes of 50 ml or higher were found to result in VaD by
one group.56 Multiple small strokes were suggested to have an additive or
multiplicative effect,9 encapsulated in the notion of multi-infarct dementia.
This is understandable, given their propensity to disrupt multiple pathways.
There is now a recognition that the location of the infarct is an important
factor, with strategically placed single-infarcts (e.g. cortical infarcts involving
the angular gyrus, inferomedial temporal lobe and medial frontal lobe, and
subcortical infarcts of the thalamus, left-sided capsular genu, caudate nuclei
308 THE AGEING BRAIN

Table 3. Risk factors for vascular dementia.

Sociodemographic
Age Increasing incidence with age, especially after 60
years
Race/ethnic Higher rates in Asian and black populations
Sex Higher rates in men
Education May have a protective effect

Atherogenic
Hypertension Major risk factor
Coronary artery disease Increases stroke risk
Diabetes mellitus Risk factor for stroke
Cigarette smoking Risk factor for stroke
Hypercholesterolaemia Risk factor for stroke
Hyperhomocystinemia Increased risk of stroke, and cognitive impair-
ment
Fibrinogen, obesity Evidence lacking

Other cardiovascular
Atrial fibrillation Risk of cerebral embolism
Mitral valve prolapse Cerebral embolism
Peripheral vascular disease Inconsistent evidence

Other factors
Genetic Weak; CADASIL an exception
Apolipoprotein E polymorphism Evidence inconsistent
Anticardiolipin antibodies Evidence inconsistent
Alcoholism Evidence inconsistent

Stroke related
Number, volume, location of stroke
Pre-existent silent infarcts
Presence of abnormal periventricular signal on magnetic resonance imaging, or (espe-
cially) on computed tomography

Note: CADASIL = cerebral autosomal dominant arteriopathy with subcortical infarct

and leukoencephalopathy.

and white matter pathways) being capable of producing severe dysfunction.

It is likely that some strategically placed infarcts lead to a disproportionate
effect on metabolism in remote areas of the brain.57 In addition to vascular
territory infarcts, infarction may occur in watershed areas. Lacunar infarcts,
usually seen in the white matter, basal ganglia, thalamus and pons, are also
recognised to produce dementia.15
The role of ischaemic white matter lesions (WMLs) in cognitive impairment
has received much recent attention. They are commonly seen on computed
tomography (CT) and especially on T2-weighted magnetic resonance imaging
(MRI) in patients with VaD, AD and normal healthy elderly individuals. Most
VASCULAR DEMENTIA 309

authors distinguish between periventricular WMLs occurring at the margins

of the lateral ventricles (“rims” and “caps”) and deep subcortical white
matter lesions sparing the U-fibres. The widespread reporting of WMLs on
MRI prompted the introduction of a new term “leukoaraiosis” (rarefaction
or thinning of the white matter).58 Their pathological significance has been
greatly debated, but when their severity had been considered, periventricular
WMLs were reported in MID to be 11.6 times higher than AD and 3.5 times
higher than normals, and subcortical WMLs were 2.6 and 13.5 times higher
respectively.59 A ‘threshold’ effect has been suggested, with cognitive impair-
ment resulting when WMLs reach a certain severity. Many of the risk factors
for WMLs are similar to those for stroke, but empirical evidence for factors
other than age and hypertension has been equivocal.60
It is debatable whether leukoaraiosis must indeed be distinguished from
Binswanger’s disease3 which has been described as a clinicopathological
entity characterised by a slow progression of a dementia associated with
psychiatric features, gait disturbance, Parkinsonism, corticobulbar features
and incontinence. It usually begins in the fifth or sixth decade and is associ-
ated with hypertension. The pathological correlate is extensive white matter
softening involving the centrum semiovale sparing U-fibres, internal capsule
and corpus callosum, usually associated with moderate cortical atrophy and
small infarcts. There is thickening and medial lipohyalinosis of the medullary
arteries of the white matter and perforators of the deep grey matter, with
hypoperfusion being hypothesised as the pathogenetic mechanism.
In addition to the arterial territory and lacunar infarcts, distal field or
watershed infarcts occur in the cortical border zones because of hypoper-
fusion. When multiple small infarcts and focal gliosis are the pathological
feature, the term granular cortical atrophy is used. Incomplete ischaemic
necrosis leads to focal areas of cell loss with reactive changes (focal gliosis)
or selective neuronal loss in cortical segments (laminar necrosis). Cortical
and deep infarcts also produce remote changes in cortical functioning. There
may be a disconnection of critical functional circuits, e.g. the frontolimbic
or thalamocortical circuits, or changes in cholinergic, noradrenergic or other
neurotransmitter functions leading to profound cognitive changes.61
CVD may lead to dementia by a process of subdural, subarachnoid or
intracerebral haemorrhage, although most authors do not include these under
the rubric of VaD.14,15 The same holds for hypoxaemic hypoxia (e.g. pure
asphyxia or respiratory failure) and histotoxic hypoxia (e.g. carbon monox-
ide or cyanide poisoning). Hypoxic-ischaemic events (e.g. seizures, cardiac
arrhythmias, pneumonia, congestive heart failure) do, however, place stroke
patients at an increased risk of dementia.
The nature of the vascular pathology is varied, with atherosclerosis, arte-
riosclerosis, lipohyalinosis, amyloid angiopathy, senile arteriolar sclerosis and
other angiopathies having been described.15 Systemic causes of thromboem-
bolism are important in some cases: inflammatory diseases (e.g. systemic
lupus erythematosis, polyarteritis nodosa, sarcoidosis, etc.), hyperviscosisty
310 THE AGEING BRAIN

syndromes (e.g. polycythemia vera, sickle cell anaemia), and embolic dis-
orders (e.g. atrial fibrillation, myocardial infarction with mural thrombus,
congential heart disease, septic, air or fat emboli).
CVD and Alzheimer-type changes not uncommonly co-occur, and 10%
to 20% of patients with dementia are classified clinically and pathologically
as having both dementias.62 In spite of this, there is no agreement on how
best to characterise a “mixed” dementia or “AD with CVD”.14,15 CVD is
known to promote the clinical expression of AD.63 Vascular factors have been
implicated in the pathogenesis of WMLs and cortical neuronal loss in AD.
Cerebral amyloid angiopathy involving pial, leptomeningeal and superficial
cortical arterioles and venules can be demonstrated in up to 90% of AD cases.
Not only can these cause severe intracranial haemorrhage, they also increase
the permeability of the blood brain barrier to serum proteins, thus interfer-
ing with neuronal metabolism. A disturbance of vascular autoregulatory
mechanisms may also contribute to selective ischaemia of the white matter.
The importance of these mechanisms is controversial, and the relationship
between CVD and AD needs further study. The pathophysiological mecha-
nisms involved in VaD are summarised in Table 4.

Subtypes of VaD and VCI

In view of the varied pathogenetic mechanisms involved, the following clas-

sification for VaD is proposed:
1. Ischaemic VaD:
1.1. Multi-infarct dementia (MID): This dementia is the consequence of
multiple cortical and cortico-subcortical infarcts in arterial territories
and distal field or watershed regions, and may be related to thrombo-

Table 4 Pathogenetic mechanisms of vascular dementia.

I. Infarct (single or multiple)

A. Arterial territory infarct
• Multiple infarcts
• Single strategic infarcts
B. Watershed infarction
C. Lacunar infarction
II. Non-infarction ischaemia
A. Subcortical leukoencephalopathy (Binswanger’s)
B. Laminar necrosis
C. Granular atrophy
D. Gliosis or sclerosis
III. Haemorrhage
A. Subdural
B. Subarachnoid
C. Intracerebral
VASCULAR DEMENTIA 311

sis in large arteries, embolic events or hypoperfusion. The presenta-

tion of this MID fits the classic description of acute onset and stepwise
progression with repeated strokes and the gradual accumulation of
pathology. The neuropsychological deficits are consistent with the
brain regions affected by infarction.
1.2. Strategic infarct dementia (SID): The clinical picture of dementia is
produced by a single strategically placed infarct. Cortical regions that
may produce this picture are hippocampal formation, angular gyrus
and cingulate gyrus. Subcortical regions include the thalamus, caudate
and globus pallidus, genu of the anterior capsule, fornix and basal
forebrain. The clinical picture will vary depending upon the lesion.
1.3. Subcortical vascular dementia (SVaD): This is characterised by lesions
in the subcortical nuclei and the white matter. Two subtypes can be
described: a) Lacunar state, with multiple lacunar infarctions, and b)
Binswanger’s disease, predominated by white matter disease which
may be associated with small infarcts. Lacunar infarcts result from the
occlusion of unbranched endarteries, usually 100-400 µm in diameter,
that result in deep cerebral infarctions, 0.2 to 15 mm3 in size, with
a mean of about 2 mm3. The arteries generally show lipohyalinosis,
often associated with fibrinoid necrosis and microatheroma. The
focal and diffuse white matter lesions are ischaemic in nature, and
are associated with fibrohyaline thickening of penetrating arteries,
and ischaemic injury of the periventricular white matter. The regions
particularly affected are frontal white matter, centrum semiovale,
and the periventricular regions (rims and caps). Pathology is char-
acterised by demyelination and incomplete infarction. Accordingly,
a prefrontal syndrome dominates the clinical picture with deficits
in frontal-executive functioning, bradyphrenia, mood disturbance,
and behavioral change. Motor and Parkinsonian symptoms are often
present. The SVaD has been suggested as a more homogeneous sub-
type of VaD that may be a candidate for drug trials.
1.4. Mixed cortical and subcortical VaD: It would not be unusual for
patients with cortical infarcts to also have extensive white matter
lesions which arguably make a major contribution to their cognitive
deficits. This is not surprising considering that the risk factors for
ischaemic leukoencephalopathy are similar to those for stroke.

2. Haemorrhagic VaD:
2.1. Intracerebral haemorrhage: ICH can lead to considerable disruption
of cerebral function and consequently dementia. It may be associated
with an infarction in many cases.
2.2. Subdural haemorrhage: In chronic cases, there may be insidious devel-
opment of cognitive deficits resulting in a picture of dementia. This is
often associated with a fluctuating course of drowsiness and mental
confusion.64
312 THE AGEING BRAIN

2.3. Subarachnoid haemorrhage: Dementia that develops acutely follow-

ing a subarachnoid haemorrhage is usually associated with other
evidence of residual brain damage.65 The delayed onset of dementia
may owe its origin to normal pressure hydrocephalus.

3. VaD with AD:

The co-occurrence of vascular and Alzheimer type changes in the brain is
a relatively common finding in dementia patients, and has been referred
to as “mixed dementia”. Since both disorders are common in the elderly,
their coincidental overlap is to be expected. However, there are important
interactions between the two pathologies. Some risk factors (e.g. hyper-
tension, cholesterol, diabetes, homocysteine) appear to be common for
the two disorders. The presence of infarction has been shown to greatly
increase the likelihood of expression of subclinical Alzheimer pathol-
ogy.66,67 About 40% of patients meeting pathological criteria for VaD
have concomitant AD pathology, and a similar proportion of AD patients
has vascular brain lesions usually considered to be coincidental.68 There
may be shared pathogenetic mechanisms such as delayed neuronal death
and apoptosis.69 This overlap has clinical as well as research implications.
In the clinical setting, the overlap may present in different ways. The cog-
nitive impairment may show a gradual onset and slow progression, which
has been generally considered to be typical of AD. However, we now rec-
ognise that subcortical VaD may have a similar longitudinal profile. Quite
often, such a presentation is punctuated by a cerebrovascular event that is
insufficient to explain all the deficits. Neuroimaging may be of assistance,
as a predominance of infarcts and WMLs sways the diagnosis in favour
of VaD. Early and progressive medial temporal atrophy would argue for a
diagnosis of VaD. It has recently been shown, however, that hippocampal
atrophy may be seen in VaD patients with no AD pathology and in the
absence of hippocampal infarction, possibly due to progressive ischaemic
necrosis.70 Functional imaging, such as SPECT, PET or fMRI may be of
assistance in the differentiation, with the typical bilateral temporo-parietal
hypometabolism or hypoperfusion being the characteristic abnormality
in AD. Neuropsychological profile may also be of some discriminating
value.71

Clinical diagnosis and prognosis

The presentation of VaD varies with the underlying pathophysiology. The
onset of MID is often sudden with a transient ischaemic attack or a stroke,
after which the clinical course may be static, remitting or progressive, with
often a fluctuating or stepwise deterioration. Predominantly subcortical
lesions may produce cognitive impairment of gradual onset and slow pro-
gression. Some other features that are more often seen in VaD, especially in
contrast with AD, are nocturnal confusion and wandering, relative preserva-
tion of emotional responsiveness and personality until the later stages of the
VASCULAR DEMENTIA 313

disease, and the presence of depression, emotional lability and incontinence

and somatic symptoms.15 A history of risk factors for CVD should alert
the clinician to the possibility of VaD, and the presence of neurological
symptoms (visual disturbances, brain stem abnormalities, sensory or motor
symptoms, etc.) and signs (hemiparesis, visual field defects, pseudobulbar
palsy, extrapyramidal signs) should provide further support. These features
are encapsulated in the Ischaemic Index,10 a score of 7 or more on which
supports the diagnosis.
The cognitive deficits in VaD are multifocal and therefore more varied than
generally seen in AD. Memory deficit, though necessary for some diagnostic
criteria, may not be as marked as that seen in AD, and verbal-performance
discrepancies are often notable. Looi and Sachdev71 surveyed the literature
on neuropsychological studies that compared VaD and AD. Of the 45 stud-
ies, 18 were excluded because of inadequacies and the remaining systemati-
cally analysed. There were a number of similarities of dysfunction between
VaD and AD. However, when matched for age, education and severity of
dementia, VaD patients had relatively superior function in verbal long-term
memory, and more impairment in frontal-executive functioning compared to
AD patients. Interpretation of the results is limited by uncertainty in diag-
nostic criteria for VaD, possible inclusion bias due to use of clinical diagnosis
alone, possible overlap of AD and VaD, and the methodological shortcom-
ings of some studies. The emphasis on frontal-executive dysfunction in VaD,
even when the diagnosis is not one of subcortical VaD is noteworthy, and has
attracted much discussion, an should arguably be incorporated in the criteria
for VaD. Other common neuropsychological deficits in VaD are visuospatial
dysfunction, dysarthria with relative preservation of language and cognitive
slowing.15 Mini-mental state examination (MMSE),23 the most commonly
used screening instrument for dementia, has a number of deficiencies when
applied to VaD: it emphasises cortical functions (language and memory) at
the expense of subcortical ones; it does not test psychomotor speed and fron-
tal function; it does not test for recognition memory; and it is insensitive to
mild cognitive impairment.
The assessment is geared toward identifying the extent of the disabilities
and all possible contributing factors. CT and/or MRI are central to the diagno-
sis and since CVD is common, guidelines are available for the topography and
severity of lesions to be considered significant. Lesions generally regarded as
trivial, such as one or two lacunes and frontal horn capping, would not meet
these criteria. At least a quarter of all white matter would need to be involved
for the lesions to be clearly significant. Functional imaging, such as single pho-
ton emission (SPECT) and positron emission tomography (PET) may provide
further information on the functional significance of any observed lesions.

Prognosis
While not being totally consistent, longitudinal studies of VaD suggest mor-
tality rates higher than those for AD and rates of admission to nursing homes
314 THE AGEING BRAIN

Table 5. Clinical assessment for vascular dementia.

• History should include onset, course and nature of cognitive deficits, and information
from the carer or other person close to the patient on subtle personality and behavioural
changes that may have been noticed.
• Full neuropsychological evaluation is required at some stage, although the Mini-Mental
State Examination, supplemented by clock-drawing and clinical assessment of frontal
lobe functioning, may be useful for screening.
• Assessment of functional losses. This may be aided by administration of scales for activi-
ties of daily living, and instrumental activities of daily living, and assessment at home
by an occupational therapist.
• Psychiatric evaluation is important, as depressive disorder is common in patients with
cerebrovascular disease and depression may produce a syndrome resembling dementia.
Anxiety disorders and psychotic symptoms may also occur in people with vascular
dementia.
• General physical examination, including pulse irregularity, cardiovascular status,
carotid bruits, fundus examination, peripheral vascular disease and hypertension (multi-
ple blood pressure measurements).
• Examination for focal neurological signs, in particular gait abnormality, visual field
defects, pseudobulbar palsy (dysarthria, dysphagia, spastic tongue, brisk jaw jerk), brisk
reflexes, extensor-plantar responses and spasticity in the limbs.
• Routine investigations, including full blood counts, erythrocyte sedimentation rate,
blood glucose, serum cholesterol and triglyceride level, syphilis serology, electrocar-
diogram, and chest x-ray. Investigations are directed towards providing evidence for
CVD and its risk factors.
• Structural brain imaging, (computed tomography or magnetic resonance imaging) is
essential to provide information on the extent, type and distribution of vascular lesions
and to exclude other potential causes of dementia, such as subdural haematoma or
tumour.
• Functional imaging, such as single photon emission tomography, positron emission
tomography and functional magnetic resonance imaging, may provide further informa-
tion on the functional significance of any observed lesions or detected abnormalities not
apparent on structural imaging.
• Other specialised investigations may include echocardiography, carotid Doppler,
antinuclear antibodies, antiphospholipid antibodies, lupus anticoagulant, serum protein
electrophoresis and cerbrospinal fluid examination.

comparable in the two. Brodaty et al.72 reported a five-year mortality rate of

63.6% (cf. 31.8% for AD) and nursing home admission rate of 31.8% (cf.
20.6% for AD). Skoog et al.73 found significantly higher three-year mortality
rates in VaD (66.6%) compared to AD (42.2%) in >85 years old subjects.
Katzman et al.74 performed a year follow-up on 320 subjects with dementia,
and using a Cox proportional hazard model, reported a mortality risk ratio
in 65–74 years old subjects of 5.4 in AD and 7.2 in VaD, compared to 2.5
in the entire cohort. These rates may be improved by better treatment and
preventative strategies.

Treatment and prevention

The management of risk factors for VaD offers the opportunity to signifi-
cantly reduce its incidence or, if dementia has already been diagnosed, halt
VASCULAR DEMENTIA 315

Table 6. Some strategies for primary prevention of vascular dementia.87

Target high-risk groups. These include elderly people; people with hypertension, dia-
betes, atrial fibrillation, or past transient ischaemic attack or stroke, and smokers.
1. Treat hypertension, optimally.
2. Treat diabetes.
3. Control hyperlipidaemia.
4. Persuade patients to cease smoking and decrease alcohol intake.
5. Prescribe anticoagulants for atrial fibrillation.
6. Provide antiplatelet therapy for high risk patients.
7. Perform carotid endarterectomy for severe (>70%) carotid stenosis.
8. Use dietary control for diabetes, obesity and hyperlipidaemia.
9. Reduce homocysteine levels in those with high levels, by folate supplementa-
tion.
10. Recommend lifestyle changes (e.g., weight loss, exercise, reduce stress, decrease
salt intake).
11. Intervene early for stroke and transient ischaemic attacks with neuroprotective
agents (e.g., propentofylline, calcium channel antagonists, N-methyl-D-aspartate
receptor antagonists, antioxidants).
12. Provide intensive rehabilitation after stroke.

its progression and sometimes achieve partial improvement. The various

strategies that could potentially be used to prevent dementia are summarised
in Table 6. Empirical data on most of these are relatively few.
The impact of control of hypertension is important and was examined in a
European study (Syst-Eur).75,76 In a follow-up over two years in patients with
systolic hypertension, active treatment was found to reduce the incidence of
dementia by 50%, from 7.7 to 3.8 cases per 1000 patient-years, which was of
borderline significance. Of the 32 cases of dementia, only two had VaD, and
23 were diagnosed with AD. The authors estimated that treatment of 1000
hypertensives for five years could prevent 19 cases of dementia. In the SHEP
trial however,77 there was no effect of the treatment of hypertension on the
incidence of dementia, even though the incidence of stroke and myocardial
infarction was reduced. In a recent double-blind comparison of losartan, an
angiotension II receptor blocker, and hydrochlorothiazide, there was a sig-
nificant improvement in cognitive function with the former but not the latter.
It was uncertain whether this was due to the direct pharmacological effects
of losartan, or because of its more effective control of blood pressure. More
work clearly needs to be done in this field to determine the effect of control of
hypertension on dementia. However, since control of hypertension is the most
important step in the prevention of strokes, its beneficial effects on health are
not in dispute. The control should start early, often decades before the antici-
pated time of emergence of cognitive deficits. In controlling hypertension, a
vigorous approach to avoid hypotension is advocated as poor autoregulation
in VaD patients increases its deleterious effects on cerebral blood flow.
Cessation of smoking has been noted to have a beneficial effect in one
study. The control of other risk factors (hyperlipidemia, platelet aggregation,
316 THE AGEING BRAIN

carotid disease) may also have a stabilising or even a beneficial effect. The
literature on the use of antiplatelet agents for the prevention of strokes is
extensive but inconclusive, and much less certain than that for heart disease.
The evidence for the efficacy of high dose aspirin (≥ 975 mg/day) in reduc-
ing the risk for stroke is convincing, but that for low dose (≤ 325 mg/day) is
equivocal, even though experimental evidence suggests that a dose as low as
40 mg/day should be an effective anti-platelet dose. Considering the gastroin-
testinal side effects of high dose aspirin, most clinicians prescribe 300–500
mg/day to VaD patients. For those “failing” aspirin therapy, other anti-plate-
let agents, such as ticlopidine, may be indicated. Anticoagulant therapy is
advocated for patients with underlying embolic diseases such as non-valvular
atrial fibrillation. Hypoxic ischaemic events, e.g. myocardial infarction, car-
diac arrhythmias, seizures, pneumonia, etc. increase the risk of dementia in
stroke patients, and should be promptly attended to.78
There is an increasing emphasis on the protection of the brain to minimise
the effect of a stroke. Most importantly, treatment strategies have been devel-
oped for early intervention, and neuroprotective agents under investigation
appear to hold great promise.79 These include NMDA receptor antagonists
such as selfotel, aptiganel and memantine, agonists at the GABA receptors,
NO-pathway inhibitors and Ca channel antagonists.
Many drugs have been investigated for the treatment of VaD but with
limited success, and interest in this field is burgeoning. Vasodilators (e.g.
hydergine and other alkaloids, cyclandelate) have some positive effects, and
modest gains in cognition have been reported with an orally active haemor-
heological agent (pentoxifylline). Other drugs that have been tried include
the vinca alkaloids (vincamine, vinburnine, vinpocetine), calcium channel
antagonists (nimodipine, nifedipine, cinnarizine, flunarizine), nootropics
(piracetam and its analogues), extracts of Ginkgo biloba and many others
(buflomedil, naftidrofuryl, idebenone, etc.), with no spectacular successes.
Some of the drugs that improve memory in some AD patients, e.g. acetyl
cholinesterase inhibitors (donepezil, rivastigmine, galantamine), may find
a role in VaD as well. Other drugs may serve a neuroprotective role, e.g.
propentofylline, calcium channel antagonists, N-methyl-D-aspartate recep-
tor antagonists, etc.
Nimodipine, a calcium channel antagonist, has attracted some attention
as a treatment of VaD. In animal models, it has been demonstrated to reduce
infarct size when administered soon after the occurrence of ischaemia. The
mechanism was suggested to be the dilatation of small and collateral cerebral
vessels,80 and reduced influx of calcium ions into depolarised neurons. 81
Two studies examining its efficacy in MID have been negative,82,83 but a
subanalysis of the latter study83 suggested a favourable response in patients
with subcortical VaD, a finding which has previously been reported84 and
which needs further examination. Memantine, a non-competitive antagonist
of NMDA receptors, has been shown in a phase III double-blind trial to be
beneficial in AD, VaD and mixed dementia.85 Propentofylline, a selective
VASCULAR DEMENTIA 317

phosphodiesterase and adenosine reuptake inhibitor, has also shown some

promise in VaD86 that is worthy of further examination.
In the absence of any drug that will unequivocally treat VaD or reverse
the cognitive decline, the mainstay of treatment is preventative and support-
ive. Supportive measures should include a rigorous treatment of psychiatric
complications such as depression, measures to facilitate independence and
community or institutional care, and support for the carer. Specific neuropsy-
chological treatment may have a role.

Conclusion

The pace of research into VaD has quickened recently but many questions
remain unanswered. In almost all areas of VaD, which include its definition,
clinical diagnosis, associated psychiatric disorders, identification of risk fac-
tors, and efficacy of interventions, more work is necessary. Since the disorder
is common, of great public health importance and potentially preventable,
such research may prove to be very cost-effective.

References

1. Dening TR, Berrios GE. The vascular dementias. In: Berrios GE, Freeman HL, edi-
tors. Alzheimer and the dementias. London: Royal Society of Medicine Services,
1991; 69–76.
2. Alzheimer A. Die arteriosklerotische atrophie des gehirns. Allg Z Psych Psy-
chisch-Gerichtlich Med. 1885; 51:809–812.
3. Binswanger O. Die Abgrenzung der allgemeine progressiven Paralyse, I–III. Berl
Klin Wochensohr. 1884; 48:1103–1105, 1137–1139, 1180–1186.
4. Alzheimer A. Die Seelenstörungen auf arteriosklerotischer Grundlage. Allg Z
Psych Psychisch-Gerichtlich Med. 1902; 59:695–701.
5. Olszewski J. Subcortical arteriosclerotic encephalopathy. World Neurol. 1962;
3:359–375.
6. Pantoni L, Garcia JH. The significance of cerebral white matter abnormalities 100
years after Binswanger’s resport: A review. Stroke. 1995; 26:1293–1301.
7. Alzheimer A. Über eine eigenartige Erkrankung der Hirnrinde. Allg Z Psych Psy-
chisch-Gerichtlich Med. 1907; 64:146–148.
8. Tomlinson BE, Blessed G, Roth M. Observations on the brains of demented old
people. J Neurol Sci. 1970; 11:205–242.
9. Hachinski VC, Lassen NA, Marshall J. Multi-infarct dementia: a cause of mental
deterioration in the elderly. Lancet. 1974; 2:207–210.
10. Hachinski VC, Iliff LD, Zilkha E, De Boulay GH, McAllister VL, Marshall J,
Russell RW, Symon L. Cerebral blood flow in dementia. Arch Neurol. 1975; 32:
632–637.
11. Tatemichi TK, Desmond DW. Epidemiology of vascular dementia. In: Prohovnik
I, Wade J, Knezevic S, Tatemichi T, editors. Vascular dementia: Current concepts.
Chichester, UK: John Wiley & Sons, 1996: 42–71.
12. Hachinski VC, Potter P, Merskey H. Leuko-araiosis. Arch Neurol. 1987; 44:
21–23.
318 THE AGEING BRAIN

13. Loeb C. Clinical criteria for the diagnosis of vascular dementia. Eur Neurol.
1988; 28:87–92.
14. Chui HC, Victoroff JI, Margolin MD, Jagust W, Shankle R, Katzman R. Criteria
for the diagnosis of ischemic vascular dementia proposed by the State of Califor-
nia Alzheimer’s Disease Diagnositc and treatment Centers. Neurology. 1992; 42:
473–480.
15. Roman GC, Tatemichi TK, Erkinjuntti T, Cummings JL, Masdeu JC, Garcia JH,
Amaducci L, Orgogozo JM, Brun A, Hoffman A. Vascular dementia: diagnostic
criteria for research studies. Report of the NINDS-AIREN international work-
shop. Neurology. 1993; 43:250–260.
16. World Health Organization. The ICD-10 classification of mental and behavioural
disorders. Diagnostic criteria for research. Geneva: World Health Organization,
1993.
17. American Psychiatric Association. Diagnostic and statistical manual of mental
disorders. 4th ed. Washington DC: American Psychiatric Association, 1994.
18. Cummings JL, Benson DF. Dementia: A clinical approach. 2nd ed. Boston: But-
terworth-Heinemann, 1992.
19. Wade JPH, Mirsen TR, Hachinski VC, Fisman M, Lau C, Merskey H. The clinical
diagnosis of Alzheimer’s disease. Arch Neurol. 1987; 44:24–29.
20. Moroney JT, Bagiella E, Desmond DW, Hachinski VC, Molsa PK, Gustafson
L, Brun A, Fischer P, Erkinjuntti T, Rosen W, Paik MC, Tatemichi TK . Meta-
analysis of the Hachinski Ischemic Score in pathologically verifed dementias.
Neurology. 1997; 49:1096–1105.
21. Drachman DA. New criteria for the diagnosis of vascula dementia. Neurology.
1993; 43:243–245.
22. Desmond DW, Tatemichi TW, Stern Y, Sano M. Cognitive function following
first stroke. Neurology. 1992; 42 (Suppl 3):426.
23. Folstein MF, Folstein SE, McHugh PR. “Mini-Mental State”: A practical method
for grading the cognitive state of patients for the clinician. J Psychiat Res. 1975;
12:189–198.
24. Wetterling T, Kanitz RD, Borgis KJ. Comparison of different diagnostic criteria
for vascular dementia (ADDTC, DSM-IV, ICD-10, NINDS-AIREN). Stroke.
1996; 27:30–36.
25. Verhey FR, Lodder J, Rozendaal N, Jolles J. Comparison of seven sets of crite-
ria used for the diagnosis of vascular dementia. Neuroepidemiology. 1996; 15:
166–172.
26. Hachinski VC. Vascular dementia: A radical redefinition. Dementia. 1994; 5:
130–132.
27. Rockwood K, Wentzel C, Hachinski V, Hogan DB, MacKnight C, McDowell I.
Prevalence and outcomes of vascular cognitive impairment. Neurology. 2000; 54:
447–451.
28. Lobo A, Launer LJ, Fratiglioni L, Andersen K, Di Carlo A, Breteler MM, Cope-
land JF, Dartigues JF, Jagger C, Martinez-Lage J, Soininen H, Hofman A. Preva-
lence of dementia and major subtypes in Europe: a collaborative study of popula-
tion-based cohorts. Neurologic diseases in the elderly research group. Neurology.
2000; 54 (Suppl 5):S4–9.
29. Hagnell O, Franck A, Grasbeck A, Ohman R, Ojesjo L, Otterbeck L, Rorsman
B. Vascular dementia in the Lundby study: 1. A prospective, epidemiological
study of incidence and risk from 1957 to 1972. Neuropsychobiology. 1992; 26:
43–49.
30. Molsa PK, Paljarvi L, Rinne JO, Rinne UK, Sako E. Validity of clinical diagnosis
in dementia: a propective clinicopathological study. J Neurol Neurosur Ps. 1985;
48:1085–1090.
VASCULAR DEMENTIA 319

31. Schoenberg BS, Kokmen E, Okazaki H. Alzheimer’s disease and other dement-
ing illnesses in a defined United States population: incidence rates and cinical
features. Ann Neurol. 1987; 22:724–729.
32. Akesson HO. A population study of senile and arteriosclerotic psychoses. Hum
Hered. 1969; 19:546–566.
33. Tatemichi TK, Paik M, Bagiella E, Desmond DW, Stern Y, Sano M, Hauser WA,
Mayeux R. Risk of dementia after stroke in a hospitalized cohort: results of a
longitudinal study. Neurology. 1994; 44:1885–1892.
34. Tatemichi TK, Desmond DW, Mayeux R, Paik M, Stern Y, Sano M, Remien RH,
Williams JB, Mohr JP, Hauser WA. Dementia after stroke: baseline frequency,
risks and clinical features in a hospitalized cohort. Neurology. 1992; 42:1185–
1193.
35. Pohjasvaara T, Erkinjuntti T, Vataja R, Kaste M. Dementia three months after
stroke: baseline frequency and effect of different definitions of dementia in the
Helsinki Stroke Aging Memory Study (SAM) cohort. Stroke. 1997; 28:785–
792.
36. Kokmen E, Whisnant JP, O’Gallon WN, Chu CP, Beard CM. Dementia after
ischemic stroke: a population-based study in Rochester, Minnesota (1960–1984).
Neurology. 1996; 46:154–159.
37. Loeb C, Gandolfo C, Croce R, Conti M. Dementia associated with lacunar infarc-
tion. Stroke. 1992; 23:1225–1229.
38. Desmond DW, Tatemichi TK, Paik M, Stern Y. Risk factors for cerebrovascular
disease as correlates of cognitive function in a stroke-free cohort. Arch Neurol.
1993; 50:162–166.
39. Gorelick PB, Brody JA, Cohen DC, Freels S, Levy P, Dollear W, Forman H,
Harris Y. Risk factors for dementia associated with mutliple cerebral infarcts: a
case-control analysis in predominantly African-American hospital-based patients.
Arch Neurol. 1993; 50:714–720.
40. Gorelick PB. Stroke prevention. Arch Neurol. 1995; 52:347–354.
41. Launer LJ, Masaki K, Petrovich H, Foley D, Havlik RJ. The association between
midlife blood pressure levels and late-life cognitive function. The Honolulu-Asia
Aging Study. J Amer Med Assoc. 1995; 274:1846–1851.
42. Skoog I, Lernfelt B, Landahl S, Palmetrz B, Andereasson L-A, Nilsson L, Pers-
son G, Oden A, Svanborg A. 15-year longitudinal study of blood pressure and
dementia. Lancet. 1996; 347:1141–1145.
43. Kilander L, Nyman H, Boberg M, Hansson L, Lithell H. Hypertension is related
to cognitivie impairment: A 20-year follow-up of 999 men. Hypertension. 1998;
31:780–786.
44. Kuusisto J, Koivisto K, Mykkanen L, Helkala E-L, Vanhanen M, Hanninen T,
Pyorala K, Riekkinen P, Laasko M. Essential hypertension and cognitive function.
The role of hyperinsulinemia. Hypertension. 1993; 21:393–417.
45. Farmer ME, Kittner SJ, Abbott RD, Wolz MM, Wolf PA, White LR. Longitudi-
nally measured blood pressure, antihypertensive medication use, and cognitive
performance: The Framingham study. J Clin Epidemiol. 1990; 43:475–480.
46. Guo Z, Viitanen M, Winblad B. Low blood pressure and 5-year mortality in a
Stockholm cohort of the very old: possible confunding by cognitive impairment
and other factors. Am J Public Health. 1997; 87:623–628.
47. Okumiya K, Matsubayashi K, Wada T, Osaki Y, doi Y, Ozawa T. J-curve relation
between blood pressure and decline in cognitive function in older people living in
community, Japan (Letter). J Am Geriat Soc. 1997; 45:1032–1033.
48. Van Swieten JC, Geyskes GG, Derix MA, Peeck BM, Ramos LMP, Van Latum
JC, Van Gijn J. Hypertension in elderly is associated with white matter lesions
and cognitive decline. Ann Neurol. 1991; 30:825–830.
320 THE AGEING BRAIN

49. Ferrer I, Bella R, Serrano M, Marti E, Guionnet N. Arteriosclerotic leucoencepha-

lopathy in the elderly and its relation to white matter lesions in Binswanger’s
disease, multi-infact encephalopathy and Alzheimer’s disease. J Neurol Sci. 1990;
98:37–50.
50. Bousser MG, Tournier-Lasserve E. Summary of the proceedings of the First Inter-
national Workshop on CADASIL. Stroke. 1994; 25:704–707.
51. Haan J, Lanser JBK, Zijderveld I, van der Does IG, Roos RA. Dementia in heredi-
tary cerebral hemorrhage with amyloidosis — Dutch type. Arch Neurol. 1990;
47:965–967.
52. Jen J, Cohen A, Yue Q, Baloh R. Hereditary systemic vasculopathy with stroke,
retinopathy and nephropathy. Neurology. 1997; 48:A328.
53. Loizou L, Jefferson J, Smith W. Subcortical arteriosclerotic encephalopathy (Bin-
swanger’s type) and cortical infarcts in a young normotensie patient. J Neurol
Neurosur Ps. 1982; 45:409–417.
54. Desmond DW. Vascular dementia: A construct in evolution. Cerebrovas Brain
Met. 1996; 8:296–325.
55. Jorm AF, Korten AE, Henderson AS. The prevalence of dementia: a quantitative
integration of the literature. Acta Psychiat Scand. 1987; 76:465–479.
56. del Ser T, Bermejo F, Portera A, Arredondo JM, Bouras C, Constantinidis J.
Vascular dementia: a clinicopathological study. J Neurol Sci. 1990; 96:1–17.
57. Mielke R, Herholz K, Grond M, Kessler J, Heiss WD. Severity of vascular demen-
tia is related to volume of metabolically impaired tissue. Arch Neurol. 1992; 49:
909–913.
58. Hachinski VC, Potter P, Merskey H. Leuko-araiosis: an ancient term for a new
problem. Can J Neurol Sci. 1986; 13:533–534.
59. Boone BK, Miller BL, Lesser IM, Mehringer CM, Hill-Gutierrez E, Goldberg MA,
Berman NG. Neuropsychological correlates of white matter lesions in healthy
elderly subjects: a threshold effect. Arch Neurol. 1992; 49:546–554.
60. Ghika J, Bogousslavsky J. White matter disease and vascular dementia. In: Wade
J, Knezevic S, Tatemichi T, Erkinjuntti T, editors. Vascular dementia: Current
concepts. Chichester, UK: John Wiley & Sons, 1996: 113–141.
61. Kataoka K,Hayakawa T, Kuroda R, Yuguchi T, Yamada K. Cholinergic deaf-
ferentation after focal cerebral infarcts in rats. Stroke. 1991; 22:1291–1296.
62. Molsa PK, Paljarvi L, Rinne JO, Rinne UK, Sako E. Validity of clinical diagnosis
in dementia: a prospective clinicopathological study. J Neurol Neurosur Ps. 1985;
48:1085–1090.
63. Snowdon DA, Greiner LH, Mortimer JA, Riley KP, Greiner PA, Markesbery WR.
Brain infarction and the clinical expression of Alzheimer disease: The Nun Study.
J Amer Med Assoc. 1997; 277:813–817.
64. Black DW. Mental changes associated with subdural haematoma. Br J Psychiat.
1984; 145:200–203.
65. Storey PB. Psychiatric sequelae of subarachnoid haemorrhage. Br Med J. 1967;
3:261–266.
66. Snowdon DA, Greiner LH, Mortimer JA, Riley KP, Greiner PA, Markesbery WR.
Brain infarction and the clinical expression of Alzheimer disease: The Nun Study.
J Amer Med Assoc. 1997; 277:813–817.
67. Heyman A, Fillenbaum GG, Welsh-Bohmer KA, Gearing M, Mirra SS, Mohs
RC, Petersen BL, Pieper CF. Cerebral infarcts in patients with autopsy-proven
Alzheimer’s disease. CERAD Part XVIII. Consortium to establish a registry for
Alzheimer’s Disease. Neurology. 1998; 51:159–162.
68. Skoog I, Kalaria RN, Breteler MMB (1999) Vascular factors and Alzheimer dis-
ease. Alz Dis Assoc Dis. 13 (Suppl 3):S106–S114.
69. Kalaria RN, Ballard C. Overlap between pathology of Alzheimer disease and
vascular dementia. Alz Dis Assoc Dis. 1999; 13 (Suppl 3):115–123.
VASCULAR DEMENTIA 321

70. Chui H. The neuropathology of vascular dementia. Presented at the inaugural

meeting of the International College of Geriatric Psychopharmacology, Hawaii,
December, 2001.
71. Looi JCL, Sachdev P. Differentiation of vascular dementia from AD on neuropsy-
chological tests. Neurology. 1999; 53:670–678.
72. Brodaty H, McGilchrist C, Harris L, Peters KE. Time until institutionalization
and death in patients with dementia: role of caregiver training and risk factors.
Arch Neurol. 1993; 50:643–650.
73. Skoog I, Lernfelt B, Landahl S, Palmertz B, Anderson LA, Svanborg A. A
population-based study of dementia in 85-year olds. N Eng J Med. 1993; 328:
153–158.
74. Katzman R, Hill LR, Yu ES, Wang ZY, Booth A, Salmon DP, Liu WT, Qu GY,
Zhang M. The malignancy of dementia. Predictors of mortality in clinically
diagnosed dementia in a population survey of Shanghai, China. Arch Neurology.
1994; 51:1220–1225.
75. Bert P, Forette F, Rigaud AS, Bouchacourt P. Traitement de l’hypertension arteri-
elle du sujet âgé: intérêt et indications. Presse Med. 1994; 23:276–280.
76. Forette F, Seuz ML, Staessen JA, Thijs L, Birkenhager WH, Barbarskiene MR,
Babeanu S, Bossini A, Gil-Extremera B, Girerd X, Laks T, Lilov E, Moisseyev
V, Tuomilehto J, Vanhanen H, Webster J, Yodfat Y, Fagard R. Prevention of
dementia in randomised double-blind placebo-controlled Systolic Hypertension
in Europe (Syst-Eur) trial. Lancet. 1998; 352:1347–1351.
77. SHEP Collaborative Research Group. Prevention of stroke by antihypertensive
drug treatment in older persons with isolated systolic hypertension. Final results
of the Systolic Hypertension in the Elderly program (SHEP). J Amer Med Assoc.
1991; 265:3255–3266.
78. Moroney JT, Bagiella E, Desmond DW, Paik MC, Stern Y, Tatemichi TK. Risk
factors for incident dementia after stroke: role of hypoxic and ischemic disorders.
Stroke. 1996; 27:1283–1289.
79. De Keyser J, Sulter G, Luiten PG. Clinical trials with neuroprotective drugs in
acute ischaemic stroke: are we doing the right thing? Trends Neurosci. 1999; 22:
535–540.
80. Auer LM, Oberbauer RW, Schalk HV. Human pial vascular reactions to intra-
venous nimodipine infusion during EC-IC bypass surgery. Stroke. 1983; 14:
210–213.
81. Dorsch NW, Branston NM, Harris RJ, Bentivoglio P, Symon L. An experimental
study of the effect of nimodipine in primate subarachnoid haemorrhage. Acta
Neurochir. 1989; 99:65–75.
82. Besson G, Bogousslavsky J. Medical treatment of acute ischemic stroke. J Cardio-
vasc Pharm. 1991; 18 (Suppl 8):S6–9.
83. Pantoni L, Bianchi C, Beneke M, Inzitari D, Wallin A, Erkinjuntti T. The Scan-
dinavian multi-infarct dementia trial: a double-blind, placebo-controlled trial on
nimodipine in multi-infarct dementia. J Neurol Sci. 2000; 175:116–123.
84. Pantoni L, Carosi M, Amigoni S, Mascalchi M, Inzitari D. A preliminary open
trial with nimodipine in patients with cognitive impairment and leukoaraiosis.
Clin Neuropharmacol. 1996; 19:497–506.
85. Gortelmeyer R, Erbler H. Memantine in treatment of mild to moderate dementia
syndrome. Drug Res. 1992; 42:904–912.
86. Kittner B. Clinical trials of propentofylline in vascular dementia. European/
Canadian Propentofylline Study Group. Alz Dis Assoc Dis. 1999; 13 (Suppl 3):
S166–S171.87.
87. Sachdev PS, Brodaty H, Looi JCL. Vascular dementia: diagnosis, management
and possible prevention. Med J Australia. 1999; 170:81–85.
Chapter 19

CONCLUDING
COMMENTS —
THE AGEING BRAIN
Perminder S Sachdev

Conclusion

When confronted with a neurological or psychiatric disorder in an elderly

individual, a clinician is likely to ask how the processes of ageing have influ-
enced the aetiology and presentation of the disorder, and will impact on its
efficient management. The clinician then seeks information about the age-
ing of the brain to make informed judgements and choices. There are many
urban myths about ageing, and some of these apply to the brain. The reviews
included in this book are an attempt to flush out these myths, and arm the cli-
nician and general researcher with the empirical facts that can be mustered to
substantiate claims about ageing. There are many salient questions: Is cogni-
tive change to be expected in an elderly individual? Is this change progressive,
relentless and unselective, or is it focal and constrained? Would every person
who lived long enough develop Alzheimer’s disease (AD)? Do our neurones
die as we grow old? What happens to the size of the brain and its metabolic
activity? How do our hormones change with age? Can anti-oxidants slow or
even stop the process of ageing? Are genes important for the ageing brain,
or is it all in the environment? How much of what we are is due to what we
eat? This book has addressed some of these questions in a language simple
enough for a general reader to understand.
Does this book also lay out a guide map for the future researcher of brain
ageing? The field is too diverse to recommend any one road to the traveller.
A cognitive neuroscientist, for example, has a very different perspective from
that of an endocrinologist when dealing with age-related processes. Genuine
324 THE AGEING BRAIN

breakthroughs will depend, however, on the successful marriage of such

diverse outlooks. The emerging role of glucocorticoids in stress, depression,
memory dysfunction, brain atrophy and ageing is one example of the power
of cross-fertilization. The research on oxidative stress in ageing has led to the
growth of a large industry extolling the virtues of antioxidants. The observa-
tions on older women have led to the examination of the role of oestrogen in
the regulation of neural plasticity. Cross-disciplinary research is more than a
mere fad in relation to the future development of this field.
Much of the progress in the future is likely to come from predictable
quarters. The hunt is on for the genes that regulate the development and
senescence of the brain. It is inevitable that this will one-day lead to their
manipulation to modify these processes, fraught though this will be with
ethical dilemmas. New technology will continue to challenge and astound
us. In the last two decades, the introduction of neuroimaging techniques has
changed the way we look at the brain. These technologies continue to evolve,
and are being complemented by developments in histopathology, such as
confocal laser microscopy, which enable us to examine single cells in great
detail. Advances in mass spectrometry have opened up the study of proteins
and led to the development of proteomics to complement the revolution in
genomics.
In medical science, many advances have come from the study of disease
processes, which may be considered nature’s joints that a researcher can
hope to carve. The quintessential disorder of ageing is AD, and its study
holds the promise of many insights about nature’s ways. Solving the riddle
of plaques and tangles may not solve the problem of ageing, but has the
potential of altering the landscape considerably. No other disorder has given
more impetus to the study of degeneration of the brain, but it may be only
the beginning. There is a limited number of ways in which brain cells lose
their vitality or viability in the aged, and some of these leave footprints such
as neuronal inclusions or inter-cellular deposits. We do not know why Lewy
bodies develop in the cortices of some individuals. We have little understand-
ing of tauopathies, and why some degeneration begins in the temporal lobes,
while others may start in the frontal, parietal or even the occipital lobes. More
than a decade after the discovery of the Huntington’s gene, we do not fully
understand the process of neuronal loss in this disease, and have no method
of stopping or modifying it. The researcher of the future will be well served
by following the footprints of nature in discovering some insights.
This book can be criticized for under-emphasizing the sociological aspects
of brain ageing. The aged brain possesses a wealth of knowledge and expe-
rience, which it holds on to even in the presence of other devastation, and
passes onto future generations. The brain also has an elegant ability to adapt
to changing circumstances. It has a built-in reserve that protects it in adver-
sity, and education, mental activity and healthy lifestyle can increase the
reserve. Nutritional and lifestyle factors may also retard the disease processes
of old age. These factors hold the potential for living well in old age, and to
CONCLUDING COMMENTS 325

stay engaged in family and society. As the demographic shift to old age occurs
in our society, there are increasing opportunities for the elderly individual and
his or her brain. While we cannot protect the brain forever, we can strive to
maximize its quality well into old age if we are to build a happy and produc-
tive future. While tabloid science abounds on this issue, it is hoped that this
book will bring empirical science to bear on any such quest.
CONTRIBUTORS
ADDRESS LIST
CONTRIBUTORS ADDRESS LIST
Carol Brayne
Department of Public Health and
Primary Care, Institute of Public
Health,
University of Cambridge,
Forvie Site, Robinson Way,
Cambridge CB2 2SR UK
Tel: 44-1223-330 334;
Fax: 44-1223-330 330
Email: carol.brayne@medschl.cam.ac.
uk
G. Anthony Broe
Senior Research Scientist, Prince of
Wales Medical Research Institute;
and
Clinical Director, Programme of
Community Health and Aged Care,
Prince of Wales Hospital;
and
Professor of Geriatric Medicine,
University of New South Wales,
Australia
Tel: +61-2-9382 4252;
Fax: +61-2-9382 4241
Email: broet@sesahs.nsw.gov.au
Janet Bryan, PhD
Research Scientist, Consumer
Science Program, CSIRO Health
Sciences and Nutrition,
Adelaide BC,
South Australia 5000,
Australia
Tel: 61-8-8303 8936;
Fax: 61-8-8303 8899
Email: janet.bryan@csiro.au
Helen Christensen, PhD
The Centre for Mental Health
Research, The Australian National
University,
Canberra ACT 0200, Australia
Tel: +61-2-6125 2741;
Fax: +61-2-6125 0733
Email: helen.christensen@anu.edu.au
329
Peter J. Crack
Centre for Functional Genomics and
Human Disease,
Monash Institute of Reproduction
and Development,
Monash University,
246 Clayton Road,
Clayton VIC 3168, Australia
Judy B. de Haan
Centre for Functional Genomics and
Human Disease,
Monash Institute of Reproduction
and Development,
Monash University,
246 Clayton Road,
Clayton VIC 3168, Australia
Tel: +61-3-9594 720;
Fax: +61-3-9594 7211
Email: judy.de.haan@med.monash.
edu.au
Geoffrey A. Donnan, MD, FRACP
Director, National Stroke Research
Institute,
Neurosciences Building,
Repat Campus, Austin &
Repatriation Medical Centre,
Heidelberg West,
Melbourne VIC 3081, Australia
Tel: +61-3-9496 2699;
Fax: +61-3-9496 2650
Email: donnan@austin.unimelb.edu.
au
Glenda M Halliday
Prince of Wales Medical Research
Institute; and
the University of New South Wales,
Sydney NSW, Australia
Mariese A Hely
Department of Neurology,
Westmead Hospital,
Westmead, Sydney, Australia
330 THE AGEING BRAIN

Paul Hertzog John B.J. Kwok

Centre for Functional Genomics and
Human Disease,
Monash Institute of Reproduction
and Development,
Monash University,
246 Clayton Road,
Clayton VIC 3168, Australia
Rocco C. Iannello
Centre for Functional Genomics and
Human Disease,
Monash Institute of Reproduction
and Development,
Monash University,
246 Clayton Road,
Clayton VIC 3168, Australia
Ismail Kola
Pharmacia Corp.,
4901 Searle Parkway,
Skokie,
Ill 60077, USA
Jillian J. Kril, PhD
Associate Professor (Geriatric Medi-
cine), Centre for Education and
Research on Ageing,
The University of Sydney;
Tel: +61-2-9767 7109;
Fax: +61-2-9767 5419
Email: jilliank@med.usyd.edu.au
Rajeev Kumar, MD, FRANZCP
Lecturer and Consultant Psychia-
trist,
The Canberra Clinical School,
University of Sydney,
PO Box 11,
Woden ACT 2606,
Australia
Email: rajeev.jumar@act.gov.au
Garvan Institute of Medical
Research,
384 Victoria Street,
Darlinghurst,
Sydney NSW 2010,
Australia
Jeffrey C.L. Looi, MBBS FRANZCP
MFPOA
Senior Specialist
Older Persons
Mental Healt Service,
6 Gaunt Place,
Gerral ACT 2005;
and
Senior Lecturer, Department of
Psychological Medicine,
Canberra Clinical School,
University of Sydney;
and
Research Fellow,
Neuropsychiatric Institute,
Prince of Wales Hospital,
Randwick, Sydney NSW, Australia
Tel: +61-2-6205 1957;
Fax: +61-2-6205 1533
Email: looij@sesahs.nsw.gov.au
Stephen R. Lord
Associate Professor,
Prince of Wales Medical Research
Institute,
Sydney NSW 2031, Australia
Tel: +61-2-9382 2721;
Fax: +61-2-9382 2722
Email: s.lord@unsw.edu.au
John G.L. Morris
Department of Neurology,
Westmead Hospital,
Westmead, Sydney NSW 2145,
Australia
Tel: 61-2-9845 6793;
Fax: 61-2-9635 6684
Email: jmorris@mail.usyd.edu.au
CONTRIBUTORS ADDRESS LIST
Perminder S. Sachdev, MD,
PhD, FRANZCP
Professor of Neuropsychiatry,
School of Psychiatry,
University of New South Wales,
Sydney,
Australia;
and
Neuropsychiatric Institute,
The Prince of Wales Hospital,
Randwick NSW 2031,
Sydney, Australia
Tel: +61-2-9382 3763;
Fax: +61-2-9382 3773/3774
Email: p.sachdev@unsw.edu.au
Peter R. Schofield, PhD, DSc
Garvan Institute of Medical
Research,
384 Victoria Street,
Darlinghurst, Sydney NSW 2010,
Australia
Tel +61-2-9295 8285;
Fax: +61-2-9295 8281
Email: p.schofield@garvan.org.au
Peter W. Schofield
Clinical Director,
Neuropsychiatry Service,
Hunter Area Health,
PO Box 833,
Newcastle NSW 2300,
Australia;
and
Conjoint Associate Professor of
Psychiatry,
University of Newcastle, Australia
Tel: 61+2+4924 6857;
Fax: 61+2+4924 6849
Email:
mdpsc@mail.newcastle.edu.au
Gary W. Small, MD
Department of Psychiatry and Biobe-
havioral Sciences,
331
The Neuropsychiatric Institute,
The Alzheimer’s Disease Center and
The Center on Aging, University of
California,
Rm. 88-201,
760 Westwood Plaza,
Los Angeles, CA, USA;
and
The VA Greater Los Angeles Health-
care System,
Los Angeles, CA, USA
Tel: 310-825-0291;
Fax: 310-825-3910
Email: gsmall@mednet.ucla.edu
George A. Smythe
Biomedical Mass Spectrometry
Facility,
Faculty of Medicine,
University of New South Wales,
Sydney NSW 2052,
Australia
Tel: +61-2-9385 2952;
Fax: +61-2-9662 4469
Email: g.smythe@unsw.edu.au
John Snowdon
Clinical Associate Professor,
University of Sydney,
PO Box 1,
Rozelle NSW 2039, Australia
Tel: 61-2-9556 9666;
Fax: 61-2-9818 5712
Email: jsnowdon@mail.usyd.edu.au
Velandai K. Srikanth,
MBBS FRACP
Research Fellow,
The Menzies Centre for Population
Health Research (University of Tas-
mania),
17 Liverpool Street,
Hobart TAS 7000, Australia
and
332
Staff Specialist Geriatrician,
Royal Hobart Hospital,
Liverpool Street,
Hobart TAS 7000,
Australia
Tel: +61-3-6226 7700;
Fax: +61-3-6226 7704
Email:
velandai.srikanth@utas.edu.au
Rebecca St George
Prince of Wales Medical Research
Institute,
Sydney NSW 2031, Australia
Tel: +61-2-9382 7916;
Fax: +61-2-9382 2722
Email: r.stgeorge@unsw.edu.au
THE AGEING BRAIN
Julian N. Trollor
Conjoint Lecturer,
School of Psychiatry,
University of New South Wales,
Sydney NSW, Australia;
and
Staff Specialist,
Neuropsychiatric Institute,
The Prince of Wales Hospital,
Sydney NSW, Australia
Tel: +61-2-9382 3755;
Fax: +61-2-9382 3774
Email: J.Trollor@unsw.edu.au
SUBJECT INDEX
SUBJECT INDEX
[Numbers in italics refer to Figures and Tables]
Activation studies
advantages, 124–125
Alzheimer’s disease, 266–268
brain patterns, 267–268
complex tasks, 127–128
deactivation of regions, 131
encoding, 128–129
frontal/executive tasks, 124
motor and sensory stimulation,
125–127
motoric tasks, 124
photic tasks, 124
purpose, 124
retrieval tasks, 129–130
spatial orientation tasks, 124, 126
summary, 130–131
verbal encoding and retrieval, 124,
129–130
visual attention tasks, 124
visual encoding and retrieval, 124
word identification tasks, 124
working memory tasks, 124
Age-associated memory impairment
(AAMI), 260
Age-related macular degeneration
(ARMD)
neurodegenerative disease, 17
Ageing
alternating fine movements, 279
balance, 276–278
brain, 4, 35
theoretical issues to consider,
35–36
brain reserve, loss of, 227
CBF and, 160–162, 161
cerebral metabolism and, 162, 163
common cause hypothesis, 85–87
depression see Depression
endocrine relationships, 140–141
eye movement, 278
facial expression, 279
gait, 276–278
health promotion industry, 8–9, 26
hormones see Hormones
intervention strategies, 69–70
longevity, increased, 15–16
longitudinal studies, 36–37, 79,
87–91, 88, 89
motor factors, 66–67
multiple pre-clinical syndromes,
19–20
335
muscle tone, 280
normal, 69, 115, 160, 249
successful, distinguished, 50
old, definition, 252
perceptions of, 3, 8
populations, 11, 12, 26, 36
demographic factors, 12
postural stability, 67, 275–276
proprioception or kinaesthesis,
65–66
regional variations in brain utilisa-
tion, 126
senses, 64–66
sensorimotor factors, relationship
with, 68
speech, 278–279
utilisation of additional networks,
126, 130
vitamin deficiency, 206–207
Ages of Mankind, 3
Alzheimer’s disease (AD)
accumulation of abnormal gene prod-
ucts, 18
activation studies, 267–268
age-associated memory impairment,
260
ageing and, 259, 324
Alzheimer’s diagnoses and reports,
299
Alzheimer’s Disease Diagnostic and
Treatment Centers (ADDTC), 302
amyloid cascade hypothesis, 174,
175, 182
amyloid precursor protein (APP)
gene, 173, 174, 265
antioxidants, 215
altering the balance of, 196, 198
apolipoprotein E (ApoE), effect of,
179, 248, 262, 267, 269
association with ageing, 4, 6, 17, 18,
35
amyloid beta (Aβ) see Amyloid beta
(Aß)
B vitamins and, 209–212
blood-brain barrier (BBB), 156
brain reserve theory, 224, 226
butyrycholinesterase (BChE) gene,
effect of, 180, 248
cerebral microvasculature, 155–156
cerebrovascular disease, co-occur-
rence with, 310
336
clinical symptoms, 174, 178, 195
Consortium to Establish a Register
for Alzheimer’s Disease (CERAD),
250
cotton wool plaques, 178
definition, 249
delayed onset, 8, 26
diagnosing, 250, 259
diagnostic categories, 260
early detection, 260
environmental factors, 27
estrogen treatment, 144
genetic research, 6, 7, 27, 173
genetic risk, 264
glucose metabolism studies, 268–
269
increasing incidence, 19, 174
intelligence, 232–233
longitudinal studies of cognitive
change
CT, 89, 90
MRI, 87, 88
low educational attainment, 228–231
mental activity, 233–234
multi-infarct dementia (MID), distin-
guished, 300, 301
neurological evaluation of at risk
groups, 123
NFT accumulation, 41, 42, 174, 178,
182, 195
NMDA receptor antagonists, use of,
316
Memantine, 316
occupational status, 229
pathologies, 251
pedigrees, 178–180, 179
perturbations of antioxidant path-
ways, 187
pre-symptomatic changes, 261
clinical trials to detect, 269–270
presenilin-1 (PS-1) gene, 173
presenilin-2 (PS-2) gene, 173
risk factors, 98
risk groups, 268, 269
spastic paraparepsis (SP), 178–180,
179
tau gene, 173, 174
vascular dementia
distinguished, 300, 304, 313–314
VaD with AD, 312
white matter lesions, 309
Amygdala
reduction in volume, 54
Alzheimer’s disease, 261–263
THE AGEING BRAIN
Amyloid angiopathy
Alzheimer’s disease, 310
cerebral microvasculature and, 155,
156
vascular pathology, 309
Amyloid beta (Aβ)
see also ß-amyloid
Alzheimer’s disease, 18, 39–40, 42,
182–183, 196
formation, 175
neurotoxic deposits, 174
tau, relationship with, 182–183
Amyloid cascade hypothesis, 174, 175,
182
Amyloid precursor protein (APP) gene
Alzheimer’s disease, 173, 174
chronic neuronal activation, 234
formation of Aß peptides, 175
mutations, 175–176
schematic diagram, 176
Amyotrophic lateral sclerosis (AML)
familial forms (FALS), 195
accumulation of free radicals,
195
antioxidant balance, altered, 195,
198
neurodegenerative disease, 17–18,
194–195
Sod1 gene, role of, 195
Angular gyrus
infarcts, 307
Antioxidants
altering the balance, 187, 190, 191,
196, 198
Alzheimer’s disease, 196, 215
cognitive ageing, 212–213
cross-sectional studies, 213–214
longitudinal studies, 214–215
lipid peroxidation in ageing, 189–
190
Parkinson’s disease, 192
pathway, 188, 188
regulation of redox status, 187
Apolipoprotein E (ApoE)
cognitive change, 84–85
dementia, subjects at risk of, 123
late onset Alzheimer’s disease
(LOAD), 179
mild cognitive impairment, predict-
ing, 123–124
Arachidonic acid (AA)
brain function and, 215–216
Arteriosclerosis
vascular pathology, 309
SUBJECT INDEX
Atherogenesis
antioxidants and, 213
Atherosclerosis
ageing and arterial changes, 154
vascular pathology, 309
white matter disease and cognitive
decline, 299, 306
Atrophy
ageing, 4, 37–38, 42, 58, 251
B vitamins, 210
brain atrophy index (BAI), 51
brain volume index (BVI), 51
CMRglu, effect on, 114
CT studies, 51
gender differences, 52
granular cortical, 309
Helsinki Aging Brain study, 52, 56
homocysteine levels, 210
longitudinal studies, 246
MRI studies, 51–52
posture, impact on, 276
resting studies, impact on, 120–121
Attention
attentional capacity, 206
nutrition and, 206
selective, 64
B vitamins
cognitive ageing and, 206–207
cross-sectional studies, 207–208
experimental studies, 208–209
longitudinal studies, 208
dementia, 209–212
memory, impact on, 208–209
Basal ganglia
infarcts, 308
lesions, 300
MRI studies, 58
N-acetylaspartate (NAA), concentra-
tion of, 131
Parkinson’s disease, 278
volume and atrophy, 53
β-amyloid
see also Amyloid beta (Aß)
accumulation, 39–40, 42, 174,
195
subtypes of plaque deposits, 40
Binswanger
disease, 276, 299, 300, 309
report on neurological degeneration,
299
Blood–brain barrier (BBB)
ageing and, 155
Alzheimer’s disease, 156
337
Blood oxygen level dependent (BOLD)
technique
activation studies, 124
fMRI scans, 99–100
Brain
ageing process, 4, 35, 58–59
theoretical issues to consider,
35–36
atrophy see Atrophy
clinicopathological studies, 41–42
exercise, 9, 84
gender differences, 37–38
gross volume, 52–54
size and neuropathological
changes, 232–233
neuronal loss, age-related, 4, 38–39,
42
nutritional factors, 9
pathology, impact of, 5–6, 41–42,
251
polyunsaturated fatty acids (PUFAs),
215–216
regional variations in utilisation by
different age groups, 126
research, current status of, 6–7
ROS-induced damage, 189
structural remodelling, 39
ventricular enlargement see Ventricu-
lar enlargement
weight, 37–38
Brain reserve hypothesis
Alzheimer’s disease, 224, 226
brain damage and, 226–227
brain reserve, definition, 224
brain size, 232–233
clinicopathological studies, 227
cognitive reserve, distinguished, 224
dementia, 224–226, 225
educational attainment, 228–231
epidemiological studies, 228
function and pathology distinguished,
223
imaging studies, 227–228
intelligence, 231–232
mental activity, 233–234
non-pathological correlates, 227
threshold concept, 224
BRAINSURF, 55
Capsular genu
infarcts, 307–308
Catalase (Cat), 188
upregulation and alteration of ratios,
189–190
338
Caudate nuclei
infarcts, 307–308
Cellular senescence
in vitro models, 190–191
in vivo models, 191–192
Cerebellum
MRI studies, 57
Cerebral autosomal dominant arteriopa-
thy with subcortical infarct and
leukoencephalopathy (CADASIL),
306–307, 308
Cerebral autosomal dominant hereditary
cerebral haemorrhage with amyloido-
sis – Dutch type, 307
Cerebral blood flow (CBF)
ageing, studies of, 160–162, 161
autoregulation, 159, 162
cerebral perfusion pressure, 158,
158
cerebral vascular resistance, 158–
159
chronic neuronal activation, 234
dementia and, 164
determinants, 157–158, 157
diet and, 206
dysautoregulation index, 162
global decline, age-related, 109
homocysteine hypothesis, 207
neuroimaging techniques, 98–100
oxygen extraction ratio (OER)
changes, 109, 114
partial volume, effect of, 109, 120–
121
regional decline, 132
regions of interest (ROI) analysis,
109
research, 157
historical background, 100–101
Technetium-HMPAO (Tc-HMPAO)
studies, 101, 106–107, 108, 121
Xenon studies, 101, 102–105, 121,
122, 157
vascular risk factors, 121–122
viscosity, 159
Cerebral metabolic rate of glucose
(CMRglu)
Alzheimer’s disease, 268–269
cerebral atrophy, 114
dementia, evaluation of at risk
groups, 123
effect of ageing, 114–115, 162
neuroimaging techniques, 98, 99,
261
regional decline, 132
THE AGEING BRAIN
resting FDG studies, 114–115, 116–
119
vascular risk factors, 121–122
Cerebral metabolic rate of oxygen
(rCMRO2)
global decline, age-related, 109, 162
neuroimaging techniques, 98
partial volume, effect of, 109
Cerebral metabolism
ageing and, 162, 163
CBF and, 160
features, 160
glucose, 160
lactate, production of, 160
normal ageing, effect of, 160
vasodilatory metabolites, 160
Cerebral perfusion pressure
CBF and, 158
idealized curve, 158
intracranial pressure (ICP), 159–160
Cerebral vascular resistance (CVR)
CBF, reduction in, 158–159
factors determining blood viscosity,
159
Cerebrovascular disease (CVD)
Alzheimer’s disease, co-occurrence
with, 310
dementia, association with, 299, 303,
309
diagnosing, 303
risk factors, 300, 313
Cerebrovascular system
ageing and arterial changes, 153–154,
162–164, 251
blood–brain barrier (BBB), 155
Atherosclerosis, 154, 299, 306
cerebral microvasculature, 154–155
Alzheimer’s disease, 155–156
disease and pathological vascular
changes, 163
focal gliosis, 309
laminar necrosis, 309
occlusive disease, 154
reactivity, 162
risk factors
CBF, effect on, 121–122
systolic hypertension, 154, 315
Cingulate
activation intensity, 267
atrophy, 53
decreased activation, 126
Cognitive ability see Cognitive speed;
Crystallised intelligence; Memory
Cognitive ageing
SUBJECT INDEX
antioxidants, 212–213
cross-sectional studies, 213–214
longitudinal studies, 214–215
definition, 76
Cognitive change, 75–76
ApoE ε2 subjects, 84–85
ApoE ε4 subjects, 84–85, 123–124
Canberra Longitudinal Study, 78–82,
81, 83
common cause hypothesis, 85–87
definition, 76
education, role of, 82
health, role of, 85
inter-individual variability,
78–85
intra-individual variability,
78–85
longitudinal studies, 76, 79–81
CT, 89, 90
MRI, 87, 88
marker variables, 82
neurophysiological approach, 75–76
non-unitary nature, 77–78
normal individuals
CT changes, 89, 90
MRI changes, 87, 88
physiological change, 85–86, 90–91
risk factor variables, 82
Seattle Longitudinal Study, 78, 80
Cognitive impairment
B vitamins, 210
homocysteine levels, 210–211
longitudinal studies of cognitive
change
CT, 89, 90
MRI, 87, 90
low educational attainment, 228–
231
mild see Mild cognitive impairment
(MCI)
synapse reduction, 4–5
Cognitive resources
definition, 206
nutrition, impact of, 206
Cognitive speed
age-related change, 79
crystallised intelligence, distin-
guished, 76
definition, 76, 206
inter-individual differences, 78–85
nutrition, impact of, 206
seven-year longitudinal change, 81
Computerised axial tomography (CT)
atrophy, 51, 53
339
Alzheimer’s disease, 261–262
cortex, 54
development of, 49
gross brain volume, 52–53
subcortical structures, 55
vascular dementia, cases of, 300,
308–309
ventricular enlargement, 51
Corpus callosum
MRI studies, 57
Cortex
CT studies, 54
MRI studies, 54–55
75, neuronal loss and Alzheimer’s dis-
ease, 174
76, problems with analysis, 55
retrieval tasks, use in, 129–130
task performance, correlation with,
126
utilisation of additional networks in
older subjects, 126, 130
Cortico-basal degeneration (CBD)
accumulation of abnormal gene prod-
ucts, 18
neurodegenerative disease, 18
Crystallised intelligence
age-related change, 78
cognitive speed, distinguished, 76
definition, 76, 206
education, protective role of, 84
inter-individual differences, 78–85
longitude individual trajectories,
83
measures and tests, 78
seven-year longitudinal change, 80
Dementia
ageing, association with, 21, 163–
164, 243, 246, 253
animal models, 246
clinical and neuropathological
studies, 243–244
incidence studies, 244–245, 245
longitudinal studies, 246
population studies, 244, 245
prevalence studies, 244–245, 245
volunteer studies, 245
B vitamins, 209–212
brain exercise, 9
brain reserve and, 224–226, 225
care and support for sufferers, 252–
253
CBF and, 164
definition, 301
340
diagnosing, 250
relaxation of diagnostic criteria,
250
epidemic, 4, 12
fronto-temporal see Fronto-temporal
dementia (FTD)
genetic profiles, 248
increasing incidence, 19–20
intelligence and, 231–232
lengthy prodromal or preclinical peri-
ods, 35–36
longevity and, 252
multiple pre-clinical syndromes as
predictors, 19–20
neurological evaluation of at risk
groups, 123
personal risk, assessment of, 246
regional variations within the brain,
6 Disinhibition-dementia-parkinsonism-
Scottish Mental Survey, 232
sex differences, 248
spastic paraparepsis (SP), 178–180,
179
stroke, following, 305
tau gene, 7
vascular see Vascular dementia
Dementia with Lewy bodies (DLB)
accumulation of abnormal gene prod-
ucts, 18, 324
neurodegenerative disease, 17, 26
Dendrites
educational attainment and complex-
ity, 230
reduction in number, 4, 39
Depression
age-bias, 284, 291
ageing and, 283, 292–293
decreasing episodes (MDEs), 286
prevalence, 284, 285, 286, 289
Composite International Diagnostic
Review (CIDI), 285
Comprehensive Psychopathological
Rating Scale, 287
cross-age studies, 284–289, 287
definition, 291–292
Diagnostic Interview Schedule, 286
DSM-III disorders, 284, 286
criterion symptoms, 292
prevalence, 288, 289–290, 290
reasons for varied survey results,
290–293
meaning, 283–284
Depressive disorders
transcranial magnetic stimulation, 8
THE AGEING BRAIN
vagus nerve stimulation, 8
Diet see Nutrition
Digit span performance
vitamin intake and, 208
Disability
ADL – impairment in personal care,
21, 24
IADL – inability to live at home with-
out domestic help, 21, 24
Kilsyth Study, 21, 22, 24
Sydney Older Persons Study, 21–25,
22
Diseases
death rates, comparison of, 20, 26
epidemiologic transition, 11–12
lengthy prodromal or preclinical peri-
ods, 35–36
systemic see Systemic diseases
amyotrophy complex, 182
Docosahexaenoic acid (DHA)
brain function and, 215–216
Dopaminergic nigro-spatial tract
Parkinson’s disease, 276, 280
Down syndrome (DS)
antioxidants, overexpression of, 187,
191, 192, 198
Sod1, 196–197
AD-like pathology, 197
Alzheimer’s pathology, 263–265
oxidative stress in early life, 197
Dutch Adult Reading Test (DART)
index of premorbid intelligence, 231
Education
protective role of, 84, 228–231,
234–235
Embolic disorders
vascular pathology, 310
Endocrine system
see also Neuroendocrinology
ageing, 140–141
androgens, 145
human growth hormone secretion
(hGH), 144–145
hypothalamic-pituitary-adrenal
(HPA) axis, 141–142, 142, 143, 145
IGF-1, 144–145
menopause, 143–144, 145
Entorhinal cortex
Alzheimer’s disease, 39, 261–262
NFTs, accumulation of, 40, 263
preservation of neuronal content, 39
Epidemiologic transition theory
SUBJECT INDEX 341

Age of degenerative diseases, 14 freezing, 277

Age of delayed degenerative diseases,
14
Age of neurodegenerative disorders,
25–27
Age of pestilence and famine, 13
Age of receding pandemics, 13
lifespan, 14–16
meaning, 11–12
population change, 13–14
systemic diseases, decline in, 18–19,
26
Exercise
brain, 9, 84
intervention strategies, 69–70
Familial multiple system tauopathy,
182
Familial progressive subcortical gliosis,
182
Fenton reaction, 188, 194
15O PET studies
complex tasks, 127–128
encoding tasks, 128–129
resting studies, 108–109, 110–113
retrieval tasks, 129–130
visual attention tests, 126
working memory, 127
Fluid intelligence
cognitive speed, distinguished, 76
definition, 76, 206
measures and tests, 77
Folate
cognitive ageing and, 206–207
cross-sectional studies, 207–208
experimental studies, 208–209
longitudinal studies, 208
memory, impact on, 209
Fronto-temporal dementia (FTD)
accumulation of abnormal gene prod-
ucts, 18
neurodegenerative disease, 17
Parkinsonism, with (FTDP-17), 182
tau gene, 180–182
Functional magnetic resonance imaging
(fMRI)
Alzheimer’s disease and vascular
dementia, distinguishing, 312
blood flow, changes in, 99–100
nature of, 99–100, 266
Gait
age-related postural stability, 67
Parkinson’s disease, 276–278
stride length, 278
Gene targeting experiments
future research, 324
in vitro models of cellular senescence,
190–191
in vivo models of cellular senescence,
191–192
Genomics see also Human genome
project
gene transfer technology, 7
Ginkgo biloba
cognitive function and, 205, 216–
217
treatment strategies for VaD, 316
Globus palladium
ageing, 4
Glucocorticoid cascade hypothesis
failure of negative feedback, 142–
143, 145
Glutamate NMDA receptor
synapses, 5
treatment strategies, 316
Glutathione peroxidase (Gpx), 188
Parkinson’s disease, 192
strokes, role in, 194
upregulation and alteration of ratios,
189–190
Gross brain volume
ageing, 58
CT studies, 52–53
MRI studies, 53–54
stroke volume and cognitive impair-
ment, 223
Haemorrhage
intracerebral (ICH), 311
subarachnoid, 312
subdural, 311
Hearing
decline in function, 65
speech comprehension, 65
Hippocampus
degeneration, 4, 39, 54
functional alteration, 4–5
neuronal loss and Alzheimer’s dis-
ease, 174, 215, 261–262
NFTs, accumulation of, 40, 263
sclerosis, 251
task performance, correlation with,
126
volume, decreasing, 54, 90
Histopathology
developments in, 324
342 THE AGEING BRAIN

Histoxic hypoxia infarcts, 307

dementia and, 309
Homocysteine hypothesis
cognitive performance and vitamin
deficiency, 207, 210–211
cross-sectional studies, 207–208
Hormones
see also Neuroendocrinology
ageing, impact on, 6, 9
androgens, 145
estrogen treatment, 144
glucocorticoid cascade hypothesis,
142–143
human growth hormone secretion
(hGH), 144–145
IGF-1, 144–145
menopause, 143–144, 145
Hounsfield unit (HU), 54, 55
Human genome project, 3, 4, 7
Hyperintensities
CBF, effect on, 122
CMRglu values, declining, 122–123
cognitive deficits, 122
increase in number, 56–57, 58
periventricular white matter (PVHs),
122
white matter (WMHs), 122–123,
261
Hypertension
Biswanger’s disease, 309
CBF, effect on, 122
stroke, 306
systolic, 154, 315
vascular dementia risk factor, 306,
312
controlling, 315
Hyperviscosisty syndromes
vascular pathology, 309–310
Hypointensities
basal ganglia, 58
Hypomethylation hypothesis
cognitive performance and vitamin
deficiency, 207
Hypothalamic–pituitary–adrenal (HPA)
axis
altered function, 141–142, 142, 143,
145
Hypoxaemic hypoxia
dementia and, 309
Immune system
link with brain, 7
lymphokines, 7
Inferomedial temporal lobe
Information processing
decline in resources, 5
Intelligence
brain reserve theory, 231–232
Interhemispheric frontal gyri
atrophy, 53
Intracranial pressure (ICP)
cerebral perfusion, 159–160
Ischaemia Scale
dementia, diagnosing, 300, 301, 302
Joints
age-related changes in sensation, 66
Leukoaraiosis
cerebral microvasculature and, 155,
156–157
Lewy body dementia
longitudinal studies, 246
Lifespan
average human, 15, 247
compression of morbidity, 14–16
demographic profile, 247–248
factors influencing, 16
longevity and dementia, 252
risk of dementia, 246–249
“survivor effect’’, 16–17, 50
Lipohyalinosis
vascular pathology, 309
Long-term potentation (LTP)
functional alteration and, 5
Longevity
education, impact of, 16–17
“survivor effect’’, 16–17, 50
Magnetic resonance imaging (MRI)
atrophy, 51–52
Alzheimer’s disease, 261–262
basal ganglia, 58
cerebellum, 57
characteristics, 58
cognitive change studies, 75
corpus callosum, 57
cortex, 54–55
development of, 49, 50
functional see Functional magnetic
resonance imaging (fMRI)
gross brain volume, 53–54
longitudinal studies of cognitive
change, 87, 90
medulla, 57
midbrain, 57
pituitary gland, 57
SUBJECT INDEX 343

pons, 57 Midbrain
subcortical structures, 55–57 MRI studies, 57
T1 – spin lattice relaxation time, Mild Cognitive Impairment (MCI)
58 diagnostic criteria, 251–252, 260
vascular dementia, cases of, 300, longitudinal studies of cognitive
308–309 change
ventricular enlargement, 51–52 CT, 89, 90
Magnetic resonance spectroscopy MRI, 87, 88
(MRS) predicting, 123–124
Alzheimer’s disease, 263–264 Mini Mental State Examination
limitations of studies, 131–132 (MMSE), 218, 245
N-acetylaspartate (NAA), concentra- antioxidant levels, 213
tion of, 100, 131 diagnosing dementia, 252
nature of, 100 problems with test, 304, 313
Medial temporal lobe (MTL) folate levels, 207, 210
decreasing volume, 54 polyunsaturated fatty acid intake,
infarcts, 307 216
Medulla Morbidity
MRI studies, 57 compression of, 14–16, 15, 26–27
Meissner corpuscles neurodegenerative diseases, 17,
age-related changes, 65 20–25
Memory Motor-neuron disease (MND)
age-associated memory impairment increasing mortality rates, 20
(AAMI), 260 late-onset, 18, 26
age-changes, 75 neurodegenerative disease, 17
antioxidants and, 213–214 Motor stimulation
brain substrates, relationship to, 90 haemodynamic response, 125
cognitive ability, 76 neural activity and BOLD signal
declarative, 77 change, 125
measures and tests, 77 signal-to-noise ratio (SNR), 125
dementia, role in diagnosis of, 301 Multi-infarct dementia (MID)
education, protective role of, 82 Alzheimer’s disease, distinguished,
encoding, 128–129 300, 301
Ginkgo biloba and, 217 calcium channel antagonists, use of,
inter-individual differences, 78–85 316
intra-individual differences, 85 Nimodipine, 316
longitude individual trajectories, 83 ischaemic vascular dementia, as,
nutrition, impact of, 206 310–311, 312
procedural, 77 onset, 312
measures and tests, 78 white matter lesions, 309
rCBF, activation and, 127 Muscles
retrieval tasks, 129–130 gait, 67, 276–278
seven-year longitudinal change, 81 quadriceps strength, 68
short-term visual, 126–127 standing, 67
stress hormones, impact of, 6 strength, 66–67
synaptic transmission changes, 5
vitamin supplements and, 208–209 Neocortex
working see Working memory NFTs, accumulation of, 40
Mental activity Neostriatum
brain reserve theory, 233–234 atrophy, 4
Mental state Neural noise
education, protective role of, 82 increased, 5
Merkel disks Neurodegenerative diseases, 17–18
age-related changes, 65 see also by name
344
accumulation of abnormal gene prod-
ucts, 18
disability statistics, 21
features, 18
gender differences, 23
Kilsyth Study, 21, 22, 24
morbidity, 17, 20–25
prevalence data, 23–24, 24
Sydney Older Persons Study, 21–25,
22
systemic diseases, replacement of, 26
under-ascertainment, 19–20
Neuroendocrinology
ageing, 140, 141–142
androgens, 145
development of field, 139
human growth hormone secretion
(hGH), 144–145
IGF-1, 144–145
menopause, 143–144, 145
proton magnetic spectroscopy, 139
Neurofibrillary tangles (NFTs)
accumulation, 40–41, 42, 174, 178,
195, 262–263
nature, 40–41
pre-symptomatic changes in Alzheim-
er’s disease, 261
timing of formation, 41
Neuroimaging
see also Computerised axial tom-
ography (CT); Magnetic resonance
imaging (MRI); Magnetic resonance
spectroscopy (MRS); Positron emis-
sion tomography (PET); Single pho-
ton emission computed tomography
(SPECT)
Alzheimer’s disease and vascular
dementia, distinguishing, 312
analysis of results, 49–50
brain reserve theory, 227–228
cognitive change studies, 75, 90
dementia, evaluation of at risk
groups, 123
FDDNP, 263
future research directions, 59, 132,
324
in vivo, 262–263
mild cognitive impairment, predict-
ing, 123–124
molecular, structural and functional
changes, 97
parameters at rest, importance of,
98
surrogate markers, using, 269–270
THE AGEING BRAIN
synthesis of findings, 58–59
technological advances, 49
validity of findings, 59
vascular dementia, cases of, 300,
308–309, 312
Neuromodulation
deficiencies, 5
Neuronal loss
cortical volume and, 4
dementia, link to, 251
Neuronal structure
diet and, 206
Neurotransmitter synthesis
diet and, 206
Nutrition
antioxidants see Antioxidants
cognitive performance and, 205–206,
324
herbal supplements, 205, 216–217
neurotransmitter synthesis, 206
vitamins see B vitamins; Folate
Omega-3
brain function and, 215–216
Pacinian corpuscles
age-related changes, 65
Pallido-ponto-nigral degeneration, 180
Parahippocampal gyrus
reduced activation, 126
volumetric decline, 54
Parieto-occipital sulcus
atrophy, 53
Parkinson’s disease (PD)
accumulation of abnormal gene prod-
ucts, 18
alternating fine movements, 279
antioxidants, 192–193, 198
association with ageing, 4, 12, 17,
18, 21, 26
balance, 276–278
brain reserve theory, 224
characteristics, 192
deep brain stimulation, 8
eye movement, 278
facial expression, 279
festination, 277
gait, 276–278
freezing, 277
stride length, 278
Gpx upregulation, effect of, 193
increasing mortality rates, 20
levodopa therapy, 278
low educational attainment, 228–
SUBJECT INDEX 345

231 regions of interest (ROI) analysis,

muscle tone, 280 268–269
perturbations of antioxidant path- resting studies, 121
ways, 187 “at risk’’ groups, 121
posture, 275–276 Precuneus
brain atrophy and, 276 decreased activation in left, 126
prevalence in the elderly, 280–281 Presenilin genes
resting tremor, 280 Alzheimer’s disease, 173
rigidity, 280 early onset (EOAD), 176–179
speech, 278–279 elevated secretion of Aβ peptides,
under-ascertainment, 19 177
Peripheral sensation mutation, 176–178
decline in function, 65–66 PS-1, 173, 177
Photic stimulation PS-2, 173
haemodynamic response, 125 Progressive supranuclear palsy (PSP)
neural activity and BOLD signal accumulation of abnormal gene prod-
change, 125 ucts, 18
signal-to-noise ratio (SNR), 125 neurodegenerative disease, 18, 182
Pituitary gland Proteases
MRI studies, 57 up-regulation, 7
Plaques Proteomics, 7–8
diffuse, 40
neuritic, 40, 262–263 Raven’s Progressive Matrices (RPM)
pre-symptomatic changes in Alzheim- complex tasks, evaluation of, 127
er’s disease, 261 Reaction time
senile count and cognitive impair- age-changes, 63–64, 68
ment, 223 Reactive oxygen species (ROS)
Polyunsaturated fatty acids (PUFAs) creation of, 187–188, 198
imbalance, 216 damage caused by, 188
role in brain function, 215–216 brain, 189
Pons stroke, 193–194
infarcts, 308 Research, 6–7
MRI studies, 57 Alzheimer’s disease, 6, 7, 27, 324
Positron emission tomography (PET) cross-disciplinary, 324
Alzheimer’s disease dementia and ageing, 243, 323
predicting, 264 animal models, 246
vascular dementia, distinguishing, clinical and neuropathological
312 studies, 243–244
basal ganglia, 58 incidence studies, 244–245, 245
cerebellum, 57 longitudinal studies, 246
cerebral atrophy, accounting for, population studies, 244, 245
120–121 prevalence studies, 244–245, 245
cerebral metabolic rate of glucose volunteer studies, 245
(rCMRglu), 98, 99, 114, 261–262 limitations, 50
cerebral metabolic rate of oxygen neuroimaging see Neuroimaging
(rCMRO2), 98 nutrition and cognitive function,
diagnostic accuracy, 265, 266 217–218
15O studies see 15O PET studies sample sizes, 50
limitations, 99 stem cell, 8
nature of, 99 Resting studies
oxygen extraction ratio (OER) see also Positron emission tomogra-
changes, 109, 114 phy (PET); Single photon emission
regional cerebral blood flow (rCBF), computed tomography (SPECT)
98, 157 “at risk’’ groups, 121
346 THE AGEING BRAIN

cerebral atrophy, accounting for, hypertension, 306

120–121 increased risk of vascular dementia,
defining healthy ageing, 115, 120 306, 315–316
resting state, definition, 120 low educational attainment, 228–
results, 121 231
screening of subjects, 115 ROS production, 194
uncontrolled mental activity, 120 Subcortical structures
Ruffini cylinders CT studies, 55
age-related changes, 65 MRI studies, 55–57
Subcortical vascular dementia (SVaD)
Single photon emission computed tom- ischaemic vascular dementia, as, 311
ography (SPECT) Substantia nigra
Alzheimer’s disease, 265 Parkinson’s disease, 192
vascular dementia, distinguishing, susceptibility to age-related disorders,
312 4
cerebral atrophy, accounting for, Sulcus
120–121 atrophy, 53, 55
diagnostic accuracy, 265 Superoxide dismutase (Sod)
index of pathology in Alzheimer’s different isoforms, distinguishing,
disease patients, 229–230 189
limitations, 99 function, 188
nature of, 98 increased levels and alteration of
regional cerebral blood flow (rCBF), ratios, 189–190
98, 157 Sod1 gene
resting studies of ageing, 106–107, overexpression, 190–191, 196
121 mutations, 195
“at risk’’ groups, 121 Synapses
Technetium-HMPAO (Tc-HMPAO) cognitive impairment, 4–5
studies, 101, 106–107, 108 long-term potentation, 5
Smell longitudinal studies, 246
decline in function, 65 reduction in number, 4, 39
Spastic paraparepsis (SP) Systemic diseases
Alzheimer’s disease, 178–180, 179 decline in, 18–19
PS-1 mutations, 178 gender differences, 23
Spino-cerebellar altrophies (SCA) neurodegenerative disorders, replace-
neurodegenerative disease, 18 ment by, 26
Standing prevalence data, 22–23, 23
age-related postural stability, 67 public health measures, success of,
sway, 67, 68 26
Stem cell research, 8 Sydney Older Persons Study, 21–25
Strategic infarct dementia (SID)
ischaemic vascular dementia, as, 311 Taste
Stress decline in function, 65
brain, impact on, 6, 9 Tau gene
environmental manipulation therapy, Aβ, relationship with, 182–183
9 Alzheimer’s disease, 173, 174, 324
hypothalamic-pituitary-adrenal dementia, 7
(HPA) axis, 141–142, 142 fronto-temporal dementia (FTD),
Stroke 180–182
antioxidant balance, altered, 194, 198 function, 182
brain reserve theory, 223–224 mutations, 180
characteristics, 193–194 exon trapping analysis, 181
dementia following, 305, 309 neurofibrillary tangles (NFTs), 40–
Gpx1, role of, 194 41, 178
SUBJECT INDEX 347

T-cells, 7 modification of risk factors, 300

Temporal cortex multi-infarct dementia (MID), 300,
preservation of neuronal content, 39 310–311, 312
Temporal horn neuroimaging, 300
increasing volume, 54 NMDA receptor antagonists, use of,
neuronal loss and Alzheimer’s dis- 316
ease, 262 Memantine, 316
Thalamus nootropics, use of, 316
ageing, 4 pathogenetic mechanisms, 309–310,
retrieval tasks, use in, 129–130 310
subcortical infarcts, 307–308 pathology, mixed, 300
task performance, correlation with, phosphodiesterase and adenosine
126 reuptake inhibitor, use of, 317
volume and atrophy, 53 prognosis, 313–314
Total creatine (Cr) risk factors, 121–122
decreasing levels, 131 age, 307, 308
Transentorhinal region gender, 307, 308
NFTs, accumulation of, 40 genetic factors, 306–307, 308
increased risk of stroke, 306, 308,
Vascular cognitive impairment (VCI) 315–316
concept of, 304 increased white matter lesions,
Vascular dementia (VaD), 19% 306, 308, 308–309
Alzheimer’s disease management of, 314–315
distinguished, 300, 304, 313–314 race, 307, 308
VaD with AD, 312 strategic infarct dementia (SID), 311
anticoagulant therapy, 316 subcortical vascular dementia
calcium channel antagonists, use of, (SVaD), 311
316 thromboembolism, 300, 309
Nimodipine, 316 treatment and prevention, 314–317
cerebral infarcts, 307 strategies, 315
location, importance of, 307 vasodilators, use of, 316
cognitive deficits, 313 vinca alkaloids, use of, 316
definition, 301–302 Ventricular enlargement
diagnosing, 301–304 ageing, 58
ADDTC criteria, 302 CT studies, 51
clinical diagnosis, 312–313, 314 gender differences, 52
DSM-IV criteria, 302 MRI studies, 51–52
ICD-10 criteria, 302 Vestibular sense
NINDS-AIREN criteria, 302–304, decline in function, 65, 68
303 gait and, 276–278
sensitivity and specificity of crite- Vibration sense
ria, 304 decline in function, 65–66, 68
education as preventative, 307 Vision
epidemiology, 304–305 decline in function, 64, 68, 126
haemorrhagic VaD, 311–312 spatial processing
intracerebral (ICH), 311 activation studies, 126
subarachnoid, 312 Vitamins see B vitamins
subdural, 311
history of research, 299–300 Wechsler Adult Intelligence Scale-
Ischaemia Scale, 300, 301, 302 Revised (WAIS-R), 77
ischaemic VaD, 310–311 White matter
mixed cortical and subcortical VaD, ageing, 58
311 CT studies, 55
NMDA receptor antagonists, use hyperintensities (WMHs), 122–123,
of, 316 300
348 THE AGEING BRAIN

infarcts, 308
lesions (WML), 277, 300
Alzheimer’s disease, 309
hypertension, 306, 312, 315
increased risk of vascular demen-
tia, 306, 308–309
multi-infarct dementia, 309
Leukoaraiosis, 156–157, 300, 309
MRI studies, 55–57
N-acetylaspartate (NAA), concentra-
tion of, 131
pallor, 251
younger persons, 53
Wisconsin Card Sort Test (WCST)
complex tasks, evaluation of, 127–
128
Working memory, 65, 127
nutrition, impact of, 206