RAPID REVIEW
Rapid review
Imaging cerebral atrophy: normal ageing to Alzheimer’s disease
Nick C Fox, Jonathan M Schott
Context With ageing populations, the prevalence of
dementia, especially Alzheimer’s disease, is set to soar.
Alzheimer’s disease is associated with progressive cerebral
atrophy, which can be seen on MRI with high resolution.
Longitudinal MRI could track disease progression and
detect neurodegenerative diseases earlier to allow prompt
and specific treatment. Such use of MRI requires accurate
understanding of how brain changes in normal ageing differ
from those in dementia.
Starting point Recently, Henry Rusinek and colleagues, in a
6-year longitudinal MRI study of initially healthy elderly
subjects, showed that an increased rate of atrophy in the
medial temporal lobe predicted future cognitive decline with
a specificity of 91% and sensitivity of 89% (Radiology 2003;
229: 691–96).
Where next? As understanding of neurodegenerative diseases increases, specific disease-modifying treatments
might become available. Serial MRI could help to determine
the efficacy of such treatments, which would be expected
to slow the rate of atrophy towards that of normal ageing,
and might also detect the onset of neurodegeneration. The
amount and pattern of excess atrophy might help to predict
the underlying pathological process, allowing specific therapies to be started. As the precision of imaging improves,
the ability to distinguish healthy ageing from degenerative
dementia should improve.
Greater numbers of us are living longer than ever before.
This increasing longevity will inevitably lead to a rise in
the number of patients with degenerative dementias such
as Alzheimer’s disease, the prevalence of which increases
exponentially after the age of 65. In addition to
devastating cognitive impairment, these disorders are
characterised by early1–3 and accelerating cerebral
atrophy.4 MRI could therefore be used to distinguish these
conditions from normal ageing. Methodological advances
in MRI now allow characterisation of the structural
changes that accompany healthy brain ageing, and might
help differentiate them from neurodegeneration.
Postmortem studies—atrophy in normal
ageing
Early evidence about the effect of ageing on brain
structure came from autopsy studies from the 19th
century. This work suggested that brain weight declined
Lancet 2004; 363: 392–94
Dementia Research Group, Institute of Neurology, Queen Square,
London WC1N 3BG, UK (N C Fox MD, J M Schott MRCP)
Correspondence to: Dr Nick C Fox
(e-mail: [email protected])
392
gradually but only slowly (about 0·1% a year) from early
adulthood until around age 60.5 Such studies were limited
by the inclusion of few elderly people, a lack of clinical
data, and were also confounded by the profound secular
changes that were taking place in the 19th and 20th
centuries in the UK: as nutrition improved, populations
had been growing taller and heavier. In the 1960s Miller
and Corsellis6 showed that the brain was not immune to
these changes: brain weights increased by about 1 g or
0·05% a year for those born between the mid-19th and
mid-20th centuries. Accounting for these changes led to
the conclusion that brain weights were stable between the
ages of 20 and 50 but fell progressively thereafter.7 Later
studies with thousands of autopsies suggested that brain
weight peaks by the mid-to-late teens and declines slowly
(0·1–0·2% a year) until age 60–70, after which losses
accelerate.8,9 An understanding of cerebral atrophy in the
individual in life had to await the advent of non-invasive
imaging.
Cross-sectional imaging in normal ageing
The introduction of computed tomography and then MRI
provided the opportunity to assess cerebral volume noninvasively and repeatedly. Early cross-sectional studies
implied that brain volume decreased and cerebrospinal
fluid volumes increased with advancing age.10–12 In
addition to global volume loss, imaging studies revealed
regional variations in the effects of ageing on the brain.
Early studies used time-consuming manual outlining of
specific structures of interest on each scan slice. Although
this approach has anatomic specificity, it is limited by the
need to make a-priori judgments as to which areas to
assess, and the technique is open to operator error and
bias.
Several more sophisticated computerised techniques
(often grouped as computational neuroanatomy13) have
since been used to provide a less biased assessment of
brain atrophy. These techniques, by negating the need for
decisions by raters about which regions of interest should
be selected and what boundaries should be chosen, allow
more accurate assessment of regional brain differences.
One such approach is voxel-based morphometry, which
involves aligning patients’ volumetric MRI scans into the
same spatial framework. These groups of scans can then
be statistically compared on a point-by-point (voxel-byvoxel) basis. Results with this technique suggested that
normal ageing is associated with a linear decline in grey
matter with relative sparing of medial temporal lobe
structures.14
A compromise between manual measures and
automated techniques involves combining an individual’s
scans in the same space and additionally using manual
delineation of the major sulci to position cortical surfaces
accurately.15,16 This technique has been used to show that
different brain regions lose volume at different rates, with
non-linearity being the rule (figure 1).15
THE LANCET • Vol 363 • January 31, 2004 • www.thelancet.com
For personal use. Only reproduce with permission from The Lancet.
RAPID REVIEW
feature of “normal” ageing, with rates of loss increasing
with comorbidity and advancing age. However, very
healthy individuals in the 60–80-year age-range have rates
of cerebral loss only slightly in excess of those in
individuals decades younger.19 Within this gradual global
cerebral decline, there are, however, regional differences
in the rates at which different brain structures atrophy.
Atrophy in neurodegenerative disorders
Figure 1: Mapping cortical change across life-span
Left lateral view of surface-rendered human brain. Scatterplots of greymatter density (y-axis) versus age 7–87 years (x-axis) show variable nonlinear decline in grey matter in cortical regions A–Y. Adapted with
permission from reference 15.
The broad conclusion from several cross-sectional
studies with various techniques is that the ageing brain
loses volume in a non-linear and region-dependent
manner, with prefrontal cortical volume declining more
rapidly than other brain regions.10,17
Cross-sectional studies are limited:
there is large inter-individual variation
in brain size and structure; the progression and rate of atrophy can only
be inferred from a one-off scan; and it
must be assumed that progression of
brain ageing is similar between individuals.
Longitudinal studies, by using
subjects as their own control, allow
progression of atrophy to be quantified
at the individual level. Within the past
few years, powerful computerised
methods have used the complexity of
the brain to match precisely (register)
an individual’s serially acquired threedimensional MRI scans to one
another.13 Cerebral atrophy can thus
be accurately assessed and localised.16,18,19
Several longitudinal imaging studies
have assessed the changes associated
with normal ageing, typically over
1–5 years. These studies, with highresolution MRI, show that rates of
global atrophy in healthy people
increase gradually with age from an
annual rate of 0·2% a year at age
30–50 to 0·3–0·5% at age 70–80.19–23
Thus, in an unbiased longitudinal
MRI study of normal ageing, while
white-matter loss was diffuse, greymatter change was more prominent in
frontal and parietal cortices than
temporal and occipital lobes.19
Despite sometimes conflicting evidence, some conclusions about
changes in brain volume can be drawn.
Brain volume loss is an inevitable
1
2
3
4
5
6
Percent decline from baseline
Longitudinal imaging in normal
ageing
There are both qualitative and quantitative differences in
rates and patterns of atrophy in neurodegeneration
compared with normal ageing. In Alzheimer’s disease,
annual rates of global brain atrophy are about 2–3%,23
compared with 0·2–0·5% in healthy controls. Silbert et al24
reported annual rates of global brain atrophy to be 2% in
Alzheimer’s disease compared with 0·4% in age-matched
controls, and found that this excess atrophy predicted the
accumulation of Alzheimer’s disease pathology at postmortem.24 Furthermore rates of brain atrophy have been
shown both to increase before the onset of symptoms
(figure 2)22–25 and to accelerate as the disease progresses.4
Different diseases with different phenotypic presentations may be associated with specific patterns of
regional atrophy. Unbiased image analysis, such as nonlinear registration of serial volumetric MRI scans, can
highlight in vivo patterns of regional atrophy that may
help distinguish these diseases from one another.16,26,27
Pathology within the medial temporal lobe has long been
0
+
1
+
2
⫺2
C
+
3
B
+
+
4
⫺4
5
+
A
6
⫺6
Symptom
onset
0
Years from first scan
4
Figure 2: Progessive atrophy in presymptomatic Alzheimer’s disease
Upper panel shows six serially acquired T1-weighted MRI scans from initially asymptomatic patient
destined to develop familial Alzheimer’s disease. Scans were acquired over 4 years before criteria for
dementia were met; the first symptoms were reported between scans 4 and 5 (arrow). Each scan
has been positionally matched (registered) to the baseline scan; red overlay represents tissue loss
compared with baseline. Lower panel plots brain volumes, derived from registered scans in upper
panel, relative to baseline (A); this gradual and accelerating loss, often difficult to see in unregistered
scans, is qualitatively and quantitatively different from global brain atrophy in (B) normal ageing and
(C) very healthy individuals with mean age of 70 years. Atrophy rates for (B) and (C) adapted from
reference 19 with permission.
THE LANCET • Vol 363 • January 31, 2004 • www.thelancet.com
For personal use. Only reproduce with permission from The Lancet.
393
RAPID REVIEW
known to be associated with memory impairment, and
pathological studies in Alzheimer’s disease have shown
prominent early involvement of medial temporal lobe
structures, especially the entorhinal cortex and hippocampus.28 Volumes of these structures can be measured
on MRI, and we now know that excess atrophy not only
predicts progression from mild memory impairment to
established Alzheimer’s disease,2,3,29–31 but also from
normal ageing to cognitive decline.1 Longitudinal studies
of apparently healthy individuals suggest that hippocampal atrophy rates increase from around 0·1–0·2% a
year in those aged 30–50, to 0·8% in those in their
mid-70s, rising further to 1·5–2% a year at 80–90.22,30,32
However, these rates are much lower than the hippocampal atrophy rates of 4–8% a year in very early
Alzheimer’s disease.22,30 Because hippocampal atrophy
accelerates several years before diagnostic criteria for
Alzheimer’s disease are fulfilled, without extended followup it is not possible to be certain that some individuals in
studies of “normal” ageing were not already in the earliest
stages of Alzheimer’s disease. In a recent such extended
longitudinal imaging study, Henry Rusinek and
colleagues1 followed up healthy subjects, aged over 60, for
6 years. Of the 45 participants completing the study, 13
showed cognitive decline and 32 remained healthy. Using
a registration-based computerised method to compare
serial MRI scans, these researchers found that atrophy in
the medial temporal lobe significantly predicted cognitive
decline. Studies such as this1 and that by Resnick et al19
are beginning to reveal the importance and time course of
increased rates of global and regional atrophy.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Imaging atrophy: the future
If specific therapies for neurodegenerative diseases
become available, serial brain imaging could be used to
assess the earliest onset of disease. Individuals at risk of
neurodegeneration on the basis of age, genetics, or family
history, could have a baseline scan which could be
compared with a follow-up scan if there was any question
of cognitive decline. Excess atrophy above that expected
for normal ageing might suggest the development of
neurodegeneration, and the regional pattern of the excess
atrophy could provide clues to the underlying disease
process. Such serial brain-imaging, possibly in combination with patterns of cognitive decline measured with
serial neuropsychometry, could permit prompt treatment
when maximum effect might be expected and provide a
means of assessing an individual’s response to therapy.
This approach has ethical, genetic, and financial
implications, which limit its routine use until diseasemodifying treatment becomes available.
We thank Jason Warren, Geoffrey Schott, and Martin Rossor for helpful
comments. NCF is an MRC Senior Clinical Fellow; JMS is an
Alzheimer’s Society (UK) Research Fellow.
References
1
2
3
4
Rusinek H, De Santi S, Frid D, et al. Regional brain atrophy rate
predicts future cognitive decline: 6-year longitudinal MR imaging
study of normal aging. Radiology 2003; 229: 691–96.
de Leon MJ, Golomb J, George AE, et al. The radiologic prediction of
Alzheimer disease: the atrophic hippocampal formation.
AJNR Am J Neuroradiol 1993; 14: 897–906.
Jack CR Jr, Petersen RC, Xu Y, et al. Rates of hippocampal atrophy
correlate with change in clinical status in aging and AD. Neurology
2000; 55: 484–89.
Chan D, Janssen JC, Whitwell JL, et al. Change in rates of cerebral
atrophy over time in early-onset Alzheimer’s disease: longitudinal MRI
study. Lancet 2003; 362: 1121–22.
394
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Blinkov SM, Glezer II. The human brain in figures and tables: a
quantitative handbook. New York: Plenum Press, 1968.
Miller AK, Corsellis JA. Evidence for a secular increase in human
brain weight during the past century. Ann Hum Biol 1977; 4: 253–57.
Miller AK, Alston RL, Corsellis JA. 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: 119–32.
Dekaban AS. Changes in brain weights during the span of human life:
relation of brain weights to body heights and body weights.
Ann Neurol 1978; 4: 345–56.
Ho KC, Roessmann U, Straumfjord JV, Monroe G. Analysis of brain
weight I: adult brain weight in relation to sex, race, and age.
Arch Pathol Lab Med 1980; 104: 635–39.
Coffey CE, Wilkinson WE, Parashos IA, et al. Quantitative cerebral
anatomy of the aging human brain: a cross-sectional study using
magnetic resonance imaging. Neurology 1992; 42: 527–36.
Jernigan TL, Press GA, Hesselink JR. Methods for measuring brain
morphologic features on magnetic resonance images: validation and
normal aging. Arch Neurol 1990; 47: 27–32.
Pfefferbaum A, Mathalon DH, Sullivan EV, et al. A quantitative
magnetic resonance imaging study of changes in brain morphology
from infancy to late adulthood. Arch Neurol 1994; 51: 874–87.
Ashburner J, Csernansky JG, Davatzikos C, et al. Computer-assisted
imaging to assess brain structure in healthy and diseased brains.
Lancet Neurol 2003; 2: 79–88.
Good CD, Johnsrude IS, Ashburner J, et al. A voxel-based
morphometric study of ageing in 465 normal adult human brains.
Neuroimage 2001; 14: 21–36.
Sowell ER, Peterson BS, Thompson PM, et al. Mapping cortical
change across the human life span. Nat Neurosci 2003; 6: 309–15.
Thompson PM, Hayashi KM, de Zubicaray G, et al. Dynamics
of gray matter loss in Alzheimer's disease. J Neurosci 2003; 23:
994–1005.
Raz N, Gunning FM, Head D, et al. Selective aging of the human
cerebral cortex observed in vivo: differential vulnerability of the
prefrontal gray matter. Cereb Cortex 1997; 7: 268–82.
Fox NC, Freeborough PA, Rossor MN. Visualisation and
quantification of rates of atrophy in Alzheimer’s disease. Lancet 1996;
348: 94–97.
Resnick SM, Pham DL, Kraut MA, et al. Longitudinal magnetic
resonance imaging studies of older adults: a shrinking brain.
J Neurosci 2003; 23: 3295–301.
Mueller EA, Moore MM, Kerr DC, et al. Brain volume preserved in
healthy elderly through the eleventh decade. Neurology 1998; 51:
1555–62.
Scahill RI, Frost C, Jenkins R, et al. A longitudinal study of brain
volume changes in normal aging using serial registered magnetic
resonance imaging. Arch Neurol 2003; 60: 989–94.
Schott JM, Fox NC, Frost C, et al. Assessing the onset of structural
change in familial Alzheimer's disease. Ann Neurol 2003; 53: 181–88.
Fox NC, Scahill RI, Crum WR, Rossor MN. Correlation between
rates of brain atrophy and cognitive decline in AD. Neurology 1999;
52: 1687–89.
Silbert LC, Quinn JF, Moore MM, et al. Changes in premorbid brain
volume predict Alzheimer’s disease pathology. Neurology 2003; 61:
487–92.
Fox NC, Warrington EK, Rossor MN. Serial magnetic resonance
imaging of cerebral atrophy in preclinical Alzheimer’s disease.
Lancet 1999; 353: 2125.
Fox NC, Crum WR, Scahill RI, et al. Imaging of onset and
progression of Alzheimer’s disease with voxel-compression mapping of
serial magnetic resonance images. Lancet 2001; 358: 201–05.
Scahill RI, Schott JM, Stevens JM, et al. Mapping the evolution of
regional atrophy in Alzheimer’s disease: unbiased analysis of
fluid-registered serial MRI. Proc Natl Acad Sci USA 2002; 99:
4703–07.
Braak H, Braak E. Neuropathological staging of Alzheimer-related
changes. Acta Neuropathol (Berl) 1991; 82: 239–59.
Fox NC, Warrington EK, Freeborough PA, et al. Presymptomatic
hippocampal atrophy in Alzheimer’s disease: a longitudinal MRI
study. Brain 1996; 119: 2001–07.
Jack CR Jr, Petersen RC, Xu Y, et al. Rate of medial temporal lobe
atrophy in typical aging and Alzheimer’s disease. Neurology 1998; 51:
993–99.
Du AT, Schuff N, Zhu XP, et al. Atrophy rates of entorhinal cortex in
AD and normal aging. Neurology 2003; 60: 481–86.
Kaye JA, Swihart T, Howieson D, et al. Volume loss of the
hippocampus and temporal lobe in healthy elderly persons destined to
develop dementia. Neurology 1997; 48: 1297–304.
THE LANCET • Vol 363 • January 31, 2004 • www.thelancet.com
For personal use. Only reproduce with permission from The Lancet.