dr. miguel Ángel villa rodríguez residencia en ... · ¿qué es el envejecimiento cerebral? n...
TRANSCRIPT
Envejecimiento cerebral
Dr. Miguel Ángel Villa Rodríguez Residencia en neuropsicología clínica
Programa de maestría y doctorado en psicología
¿Qué es el envejecimiento cerebral?
n “El envejecimiento es una serie de cambios morfológicos, fisiológicos y metabólicos que ocurren en los tejidos vivos con el paso del tiempo, que no resultan de enfermedad o agentes extrínsecos y que inevitablemente acercan al individuo a la muerte.
n Los cambios cerebrales que más frecuentemente ocurren en el envejecimiento normal son: disminución del peso y volumen cerebrales, atrofia cortical, pérdida de neuronas corticales y de algunos núcleos subcorticales, aumento de gránulos de lipofuscina en neuronas y glía, cambios hipertróficos en la glía astrocitaria; las estructuras filogenéticas y ontogenéticamente más antiguas son las primeras afectadas.”
n Escobar-Izquierdo, A. (2001). Envejecimiento cerebral normal. Revista Mexicana de Neurociencia, 2(4), 197-202.
3
Escobar et al. (1963)
13451306
1229
1146
10001050110011501200125013001350
20-40 años 60-70 años
Disminución del peso encefálico(n=655 encéfalos)
Hombres Mujeres
Lamasacerebralsereduce
Lacortezaseadelgaza
Disminuyelasubstanciablanca
Disminuyelaproduccióndeneurotransmisores
6
La masa cerebral se reduce
A Longitudinal Study of Brain Volume Changesin Normal Aging Using Serial Registered MagneticResonance ImagingRachael I. Scahill, MA; Chris Frost, MA, DipStat; Rhian Jenkins, BSc; Jennifer L. Whitwell, BA;Martin N. Rossor, MD, FRCP; Nick C. Fox, MD, MRCP
Objective: To investigate the effect of age on global andregional brain volumes and rates of atrophy, and to com-pare directly results based on cross-sectional and longi-tudinal data.
Methods: Thirty-nine healthy control subjects (agerange, 31-84 years) underwent serial magnetic reso-nance imaging assessments. Measurements included thewhole-brain, temporal lobe, hippocampal, and ventricu-lar volumes at baseline and for repeat scans.
Results: We found significant decreases in cross-sectional whole-brain (P!.001), temporal lobe (P!.001),and hippocampal (P=.003) volumes and a significant in-crease in ventricular volume (P!.001) with increasingage. Cross-sectional and longitudinal estimates of atro-
phy rates were similar. We also found directional evi-dence of acceleration in atrophy rates with increasing agein all analyses, with the most marked changes occurringafter 70 years of age. This increase in rates after 70 yearsof age was particularly marked in the ventricles (P!.001)and the hippocampi (P=.01).
Conclusions: We found a significant age-associated de-crease in global and regional brain volumes. Some evi-dence indicates that this decline in brain volumes maybe due to a nonlinear acceleration in rates of atrophy withincreasing age. A better understanding of this process mayhelp to discriminate normal age-related changes from neu-rodegenerative diseases.
Arch Neurol. 2003;60:989-994
T HE INCREASES in life expec-tancy during the past 100years constitute one of thegreat achievements ofmedical research and pub-
lic health. However, the aging brain is as-sociated with a greatly increased preva-lence of dementia. Addressing this publichealth problem is a major challenge cur-rently facing medical research.
Magnetic resonance imaging (MRI)–based measurements of the brain have beenproposed as aids in the diagnosis of Alz-heimer disease (AD) and other types of de-mentia. These measures have shown globalbrain atrophy,1-3 reduced temporal lobeand in particular hippocampal vol-umes,4-8 and increased cerebrospinal fluid(CSF) spaces9 when subjects with AD arecompared with control subjects. Rates ofregional or global atrophy have also beenproposed as surrogate markers of diseaseprogression for use in clinical trials.10 Fordiagnostic purposes, it is important to un-derstand the structural aging process to dif-ferentiate pathologic rates of atrophy fromnormal age-related changes. For trial pur-poses, the best one could expect of a drug
designed to slow disease progression in ADis to reduce the rate of atrophy to that seenin normal aging. As a result, normative lon-gitudinal data are needed for sample sizecalculations for progression trials.
Numerous cross-sectional imagingstudies have found a correlation betweenincreasing age and decreasing brain vol-umes,11-18 and these findings are substan-tiated by postmortem data.19,20 Some stud-ies have shown age-related decreases inhippocampal,18,21,22 temporal,13,22 and fron-tal lobe16,23-25 volumes and increases in CSFspaces.9,11,12,14,16-18 However, some contra-dictory evidence from other studies showsno significant association between age andventricular CSF spaces13,14 and whole-brain,21 temporal lobe,21,26 or hippocam-pal27,28 volume.
Cross-sectional studies have a num-ber of disadvantages. First, the largeamount of between-individual variationthat exists in the normal cerebral mor-phology reduces the sensitivity of meth-ods to detect true cerebral volume differ-ences between groups of subjects ofdifferent ages. For example, a large bio-logical variation exists in head size be-
ORIGINAL CONTRIBUTION
From the Dementia ResearchGroup, Department of ClinicalNeurology, Institute ofNeurology, University CollegeLondon (Mss Scahill, Jenkins,and Whitwell; Mr Frost; andDrs Rossor and Fox); theMedical Statistics Unit, LondonSchool of Hygiene and TropicalMedicine (Mr Frost); and theDivision of Neuroscience andPsychological Medicine,Faculty of Medicine, ImperialCollege of Science, Engineeringand Medicine (Dr Rossor),London, England.
(REPRINTED) ARCH NEUROL / VOL 60, JULY 2003 WWW.ARCHNEUROL.COM989
©2003 American Medical Association. All rights reserved.Downloaded From: http://archneur.jamanetwork.com/ by a SUNY Binghamton User on 05/13/2015
Date of download: 2/14/2016" Copyright © 2016 American Medical Association. All rights reserved."
From: A Longitudinal Study of Brain Volume Changes in Normal Aging Using Serial Registered Magnetic Resonance Imaging!
Arch Neurol. 2003;60(7):989-994. doi:10.1001/archneur.60.7.989"
dinal results did not show a significant, consistentcontinuous age-related acceleration in the rates of globalor regional brain atrophy, which concurs with the find-ings of previous longitudinal studies.13,14 However, a sig-nificant acceleration in ventricular expansion with agewas observed, as in recent work.14 Although the studyby Resnick et al14 was based on much older subjects, therates of ventricular enlargement we report in the sub-jects older than 60 years are entirely consistent with thoseof that much larger study. Nevertheless, ventricular ex-
pansion may not be wholly attributable to brain atro-phy. Other factors, such as CSF dynamics, change therelative distribution of CSF in ventricular and sulcal spacesand thereby may influence these observations.
Although we did not find a consistent accelerationin rates across the whole adult age range, our results sug-gest a nonlinear pattern, with rates remaining compar-atively stable before 70 years of age, but accelerating there-after. The small sample sizes make it difficult to determinethe point at which this acceleration occurs. In addition,
1 350 000
1 300 000
1 250 000
1 200 000
1 150 000
1 100 000
1 050 000
1 000 000
60 000
50 000
40 000
30 000
20 000
10 000
0
150 000
145 000
140 000
135 000
130 000
125 000
120 000
115 000
110 000
105 000
100 000
30 40 50 60 70 80
Who
le B
rain
Vol
ume,
mm
3
30 40 50 60 70 80
Vent
ricul
ar
Volu
me,
mm
3
30 40 50 60 70 80
Tota
l Tem
pora
l Lob
e V
olum
e, m
m3
6500
4500
6000
5500
5000
400030 40 50 60 70 80
Age, y
Tota
l Hip
poca
mpa
l V
olum
e, m
m3
Rate
of B
rain
Atro
phy
per Y
ear,
%Ra
te o
f Ven
tricu
lar
Enl
arge
men
t, m
m3 /
yRa
te o
f Tem
pora
l Lob
e A
troph
y pe
r Yea
r, %
Rate
of H
ippo
cam
pal
Atro
phy
per Y
ear,
%
2.5
1.5
2.0
1.0
0.5
–0.5
0.0
–1.0
–1.530 40 50 60 70 80
6000
4000
5000
3000
2000
1000
030 40 50 60 70 80
3.0
1.5
2.5
2.0
1.0
0.5
0.0
–0.530 40 50 60 70 80
4.0
0.0
3.0
2.0
1.0
0.5
3.5
2.5
1.5
–1.0
–0.5
30 40 50 60 70 80Year
A B
Total intracranial volume–adjusted cross-sectional volumes (A), and longitudinal rates of change (B) in the volume of the whole brain, ventricles, temporal lobes, andhippocampi across ages ranging from 30 to 84 years. Mean values with 95% confidence intervals are plotted against the mean age for each of the 5 age groups.
(REPRINTED) ARCH NEUROL / VOL 60, JULY 2003 WWW.ARCHNEUROL.COM992
©2003 American Medical Association. All rights reserved.Downloaded From: http://archneur.jamanetwork.com/ by a SUNY Binghamton User on 05/13/2015
Volumen total
Volumen de los ventrículos
Volumen Lob temp
Volumen hipocampo
Date of download: 2/14/2016" Copyright © 2016 American Medical Association. All rights reserved."
From: A Longitudinal Study of Brain Volume Changes in Normal Aging Using Serial Registered Magnetic Resonance Imaging!
Arch Neurol. 2003;60(7):989-994. doi:10.1001/archneur.60.7.989"
This illustration shows a three-dimensional side view of one of two cerebral hemispheres of the brain. To help visualize this, imagine looking at the cut side of an avocado sliced long ways in half, with the pit still in the fruit. In this illustration, the “pit” is several key structures that lie deep within the brain (the hypothalamus, amygdala, and hippocampus) and the brain stem.
Side View of the Brain
Raz, N. (2005). The aging brain observed in vivo. Differential changes and their modifiers. En: Cabeza, R., Nyberg, L. y Park, D. (Eds.). Cognitive neuroscience of aging. Nueva York: Oxford University Press.
Associations between age and gray matter volume inanatomical brain networks in middle-aged to older adults
Anne Hafkemeijer,1,2,3 Irmhild Altmann-Schneider,2,4
Anton J. M. de Craen,4,5 P. Eline Slagboom,4,6
Jeroen van der Grond2 and Serge A. R. B. Rombouts1,2,3
1Institute of Psychology, Leiden University, 2Department of Radiology, LeidenUniversity Medical Center, 3Leiden Institute for Brain and Cognition, LeidenUniversity, 4 Department of Molecular Epidemiology, NetherlandsConsortium for Healthy Ageing, Leiden University Medical Center,5Department of Gerontology and Geriatrics, Leiden University MedicalCenter, 6Department of Molecular Epidemiology, Leiden University MedicalCenter, Leiden, The Netherlands
Summary
Aging is associated with cognitive decline, diminished brain
function, regional brain atrophy, and disrupted structural and
functional brain connectivity. Understanding brain networks in
aging is essential, as brain function depends on large-scale
distributed networks. Little is known of structural covariance
networks to study inter-regional gray matter anatomical associ-
ations in aging. Here, we investigate anatomical brain networks
based on structural covariance of gray matter volume among 370
middle-aged to older adults of 45–85 years. For each of 370
subjects, we acquired a T1-weighted anatomical MRI scan. After
segmentation of structural MRI scans, nine anatomical networks
were defined based on structural covariance of gray matter
volume among subjects. We analyzed associations between age
and gray matter volume in anatomical networks using linear
regression analyses. Age was negatively associated with gray
matter volume in four anatomical networks (P < 0.001, cor-
rected): a subcortical network, sensorimotor network, posterior
cingulate network, and an anterior cingulate network. Age was
not significantly associated with gray matter volume in five
networks: temporal network, auditory network, and three cere-
bellar networks. These results were independent of gender and
white matter hyperintensities. Gray matter volume decreases
with age in networks containing subcortical structures, sensori-
motor structures, posterior, and anterior cingulate cortices. Gray
matter volume in temporal, auditory, and cerebellar networks
remains relatively unaffected with advancing age.
Key words: aging; atrophy; brain; gray matter; magnetic
resonance imaging; structural covariance networks.
Abbreviations
FLAIR fluid-attenuated inversion recovery
FSL functional magnetic resonance imaging of the brain
software library
ICA independent component analysis
MNI Montreal Neurological Institute
WMHs white matter hyperintensities
Introduction
It is well recognized that the process of aging is associated with
cognitive decline and diminished brain function (Grady, 2012). In
addition, numerous neuroimaging studies have unequivocally shown
that aging is associated with loss of brain tissue, in which process
especially the gray matter seems affected. Volumetric and morphomet-
ric neuroimaging studies have demonstrated a consistent age-depen-
dent decrease in regional gray matter volume, mainly expressed in the
temporal lobe and hippocampus, the cingulate cortex, and prefrontal
regions (Good et al., 2001; Jernigan et al., 2001; Resnick et al., 2003;
Raz et al., 2005).
There is increasing evidence that, in addition to brain atrophy, aging
and loss of cognitive function at high ages are associated with disrupted
structural and functional brain connectivity. It has been shown that
functional connectivity decreases with age, especially connectivity in the
default mode network between the medial prefrontal cortex, anterior
and posterior cingulate cortex, precuneus, parietal cortex, and hippo-
campus (Damoiseaux et al., 2008; Hafkemeijer et al., 2012; Ferreira &
Busatto, 2013). Furthermore, aging is associated with disrupted white
matter anatomical connections, specifically in the frontal white matter,
anterior cingulum, and the genu of the corpus callosum (Salat et al.,
2005; Madden et al., 2012).
In addition to functional brain networks and white matter anatomical
connectivity, population (intersubject) covariance of gray matter volume
can be used to study inter-regional anatomical associations (Alexander-
Bloch et al., 2013). The integrity of these gray matter structural
covariance networks changes throughout lifespan (Wu et al., 2012,
2013). Here, we will investigate the integrity of gray matter anatomical
networks in the aging brain. In this respect, mainly the structural
covariance of the default mode network has been studied, showing a
breakdown with increasing age (Spreng & Turner, 2013). While most
studies focused on the default mode network, there is evidence for age-
dependent decreases in other anatomical networks (Montembeault
et al., 2012; Segall et al., 2012; Li et al., 2013).
Currently, anatomical networks are mostly studied using a model-
driven seed-based approach with a priori hypotheses of manually selected
regions of interest and their connected networks (Montembeault et al.,
2012; Zielinski et al., 2012; Li et al., 2013; Soriano-Mas et al., 2013). The
manual selection of regions of interest might introduce a selection bias
(Damoiseaux & Greicius, 2009). To avoid this, we will use a model-free
method to investigatewhole-brain anatomical networks in an unrestricted
exploratory way. This method has proven to be a powerful tool to
characterize structural networks in schizophrenia (Xu et al., 2009). Here,
we will apply this method to study gray matter anatomical networks in
middle-aged to older adults.
In this study, we explored anatomical networks in a large group of
middle-aged to older adults (45–85 years, n = 370). Our aim was to
investigate whole-brain anatomical networks to explore which networks
are associated with the process of healthy aging and which networks do
not show an age association.
Correspondence
Anne Hafkemeijer, Department of Radiology, Leiden University Medical Center,
Postzone C2-S, PO Box 9600, 2300 RC Leiden, The Netherlands.
Tel.: +31 71 526 3998; e-mail: [email protected]
Accepted for publication 21 August 2014
1068 ª 2014 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.This is an open access article under the terms of the Creative Commons Attribution License, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited.
Aging Cell (2014) 13, pp1068–1074 Doi: 10.1111/acel.12271
Agin
g Ce
ll
Associations between age and gray matter volume inanatomical brain networks in middle-aged to older adults
Anne Hafkemeijer,1,2,3 Irmhild Altmann-Schneider,2,4
Anton J. M. de Craen,4,5 P. Eline Slagboom,4,6
Jeroen van der Grond2 and Serge A. R. B. Rombouts1,2,3
1Institute of Psychology, Leiden University, 2Department of Radiology, LeidenUniversity Medical Center, 3Leiden Institute for Brain and Cognition, LeidenUniversity, 4 Department of Molecular Epidemiology, NetherlandsConsortium for Healthy Ageing, Leiden University Medical Center,5Department of Gerontology and Geriatrics, Leiden University MedicalCenter, 6Department of Molecular Epidemiology, Leiden University MedicalCenter, Leiden, The Netherlands
Summary
Aging is associated with cognitive decline, diminished brain
function, regional brain atrophy, and disrupted structural and
functional brain connectivity. Understanding brain networks in
aging is essential, as brain function depends on large-scale
distributed networks. Little is known of structural covariance
networks to study inter-regional gray matter anatomical associ-
ations in aging. Here, we investigate anatomical brain networks
based on structural covariance of gray matter volume among 370
middle-aged to older adults of 45–85 years. For each of 370
subjects, we acquired a T1-weighted anatomical MRI scan. After
segmentation of structural MRI scans, nine anatomical networks
were defined based on structural covariance of gray matter
volume among subjects. We analyzed associations between age
and gray matter volume in anatomical networks using linear
regression analyses. Age was negatively associated with gray
matter volume in four anatomical networks (P < 0.001, cor-
rected): a subcortical network, sensorimotor network, posterior
cingulate network, and an anterior cingulate network. Age was
not significantly associated with gray matter volume in five
networks: temporal network, auditory network, and three cere-
bellar networks. These results were independent of gender and
white matter hyperintensities. Gray matter volume decreases
with age in networks containing subcortical structures, sensori-
motor structures, posterior, and anterior cingulate cortices. Gray
matter volume in temporal, auditory, and cerebellar networks
remains relatively unaffected with advancing age.
Key words: aging; atrophy; brain; gray matter; magnetic
resonance imaging; structural covariance networks.
Abbreviations
FLAIR fluid-attenuated inversion recovery
FSL functional magnetic resonance imaging of the brain
software library
ICA independent component analysis
MNI Montreal Neurological Institute
WMHs white matter hyperintensities
Introduction
It is well recognized that the process of aging is associated with
cognitive decline and diminished brain function (Grady, 2012). In
addition, numerous neuroimaging studies have unequivocally shown
that aging is associated with loss of brain tissue, in which process
especially the gray matter seems affected. Volumetric and morphomet-
ric neuroimaging studies have demonstrated a consistent age-depen-
dent decrease in regional gray matter volume, mainly expressed in the
temporal lobe and hippocampus, the cingulate cortex, and prefrontal
regions (Good et al., 2001; Jernigan et al., 2001; Resnick et al., 2003;
Raz et al., 2005).
There is increasing evidence that, in addition to brain atrophy, aging
and loss of cognitive function at high ages are associated with disrupted
structural and functional brain connectivity. It has been shown that
functional connectivity decreases with age, especially connectivity in the
default mode network between the medial prefrontal cortex, anterior
and posterior cingulate cortex, precuneus, parietal cortex, and hippo-
campus (Damoiseaux et al., 2008; Hafkemeijer et al., 2012; Ferreira &
Busatto, 2013). Furthermore, aging is associated with disrupted white
matter anatomical connections, specifically in the frontal white matter,
anterior cingulum, and the genu of the corpus callosum (Salat et al.,
2005; Madden et al., 2012).
In addition to functional brain networks and white matter anatomical
connectivity, population (intersubject) covariance of gray matter volume
can be used to study inter-regional anatomical associations (Alexander-
Bloch et al., 2013). The integrity of these gray matter structural
covariance networks changes throughout lifespan (Wu et al., 2012,
2013). Here, we will investigate the integrity of gray matter anatomical
networks in the aging brain. In this respect, mainly the structural
covariance of the default mode network has been studied, showing a
breakdown with increasing age (Spreng & Turner, 2013). While most
studies focused on the default mode network, there is evidence for age-
dependent decreases in other anatomical networks (Montembeault
et al., 2012; Segall et al., 2012; Li et al., 2013).
Currently, anatomical networks are mostly studied using a model-
driven seed-based approach with a priori hypotheses of manually selected
regions of interest and their connected networks (Montembeault et al.,
2012; Zielinski et al., 2012; Li et al., 2013; Soriano-Mas et al., 2013). The
manual selection of regions of interest might introduce a selection bias
(Damoiseaux & Greicius, 2009). To avoid this, we will use a model-free
method to investigatewhole-brain anatomical networks in an unrestricted
exploratory way. This method has proven to be a powerful tool to
characterize structural networks in schizophrenia (Xu et al., 2009). Here,
we will apply this method to study gray matter anatomical networks in
middle-aged to older adults.
In this study, we explored anatomical networks in a large group of
middle-aged to older adults (45–85 years, n = 370). Our aim was to
investigate whole-brain anatomical networks to explore which networks
are associated with the process of healthy aging and which networks do
not show an age association.
Correspondence
Anne Hafkemeijer, Department of Radiology, Leiden University Medical Center,
Postzone C2-S, PO Box 9600, 2300 RC Leiden, The Netherlands.
Tel.: +31 71 526 3998; e-mail: [email protected]
Accepted for publication 21 August 2014
1068 ª 2014 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.This is an open access article under the terms of the Creative Commons Attribution License, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited.
Aging Cell (2014) 13, pp1068–1074 Doi: 10.1111/acel.12271
Agin
g Ce
ll
n=370; 45-85 años
showed that this anatomical network changes with age in healthy and
pathological aging (Spreng & Turner, 2013). Visual inspection of our
data showed spatial overlap between the structures of this anatomical
network and the default mode functional connectivity network found in
other studies (Beckmann et al., 2005; Damoiseaux et al., 2006; Laird
et al., 2011). The default mode network is affected by age-related
atrophy (Buckner et al., 2008) and age-related decreases in functional
connectivity (Damoiseaux et al., 2008; Hafkemeijer et al., 2012; Ferreira
& Busatto, 2013).
The anatomical network containing predominantly the anterior
cingulate cortex (network d) shows spatial overlap with a functional
connectivity network associated with executive control functions
(Beckmann et al., 2005; Damoiseaux et al., 2006; Laird et al., 2011).
The associations between age and gray matter volume in this network is
supported by other anatomical network studies (Bergfield et al.,
2010; Montembeault et al., 2012). It has been suggested that the
age-dependent breakdown of this network may explain the difficulties in
cognitively demanding tasks generally observed in elderly (Montembeault
et al., 2012).
Prior studies mostly focused on age-related differences in the aging
brain. Relatively few studies have sought to identify anatomical networks
that were not associated with age. Functional connectivity in somato-
sensory and cerebellar networks does not show an association with
advancing age (Tomasi & Volkow, 2012). Here, we showed that gray
(B)(A)
a
b
c
d
e
f
g
h
i
Fig. 1 Gray matter structural networksand associations with age. (A) Overview ofthe nine anatomical networks based on thecovariation of gray matter volumes amongmiddle-aged to older adults. Networks areoverlaid on the most informative coronal,sagittal, and transverse slices of the MNI-152 standard anatomical image. (B) Theassociation between age and gray mattervolume in the anatomical networks isillustrated by bar graphs. Error bars indicatethe standard error of the mean. Age wasnegatively associated with gray mattervolume in network a–d and was notsignificantly associated with gray mattervolume in network e–i.
Anatomical networks in middle-aged to older adults, A. Hafkemeijer et al.1070
ª 2014 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
showed that this anatomical network changes with age in healthy and
pathological aging (Spreng & Turner, 2013). Visual inspection of our
data showed spatial overlap between the structures of this anatomical
network and the default mode functional connectivity network found in
other studies (Beckmann et al., 2005; Damoiseaux et al., 2006; Laird
et al., 2011). The default mode network is affected by age-related
atrophy (Buckner et al., 2008) and age-related decreases in functional
connectivity (Damoiseaux et al., 2008; Hafkemeijer et al., 2012; Ferreira
& Busatto, 2013).
The anatomical network containing predominantly the anterior
cingulate cortex (network d) shows spatial overlap with a functional
connectivity network associated with executive control functions
(Beckmann et al., 2005; Damoiseaux et al., 2006; Laird et al., 2011).
The associations between age and gray matter volume in this network is
supported by other anatomical network studies (Bergfield et al.,
2010; Montembeault et al., 2012). It has been suggested that the
age-dependent breakdown of this network may explain the difficulties in
cognitively demanding tasks generally observed in elderly (Montembeault
et al., 2012).
Prior studies mostly focused on age-related differences in the aging
brain. Relatively few studies have sought to identify anatomical networks
that were not associated with age. Functional connectivity in somato-
sensory and cerebellar networks does not show an association with
advancing age (Tomasi & Volkow, 2012). Here, we showed that gray
(B)(A)
a
b
c
d
e
f
g
h
i
Fig. 1 Gray matter structural networksand associations with age. (A) Overview ofthe nine anatomical networks based on thecovariation of gray matter volumes amongmiddle-aged to older adults. Networks areoverlaid on the most informative coronal,sagittal, and transverse slices of the MNI-152 standard anatomical image. (B) Theassociation between age and gray mattervolume in the anatomical networks isillustrated by bar graphs. Error bars indicatethe standard error of the mean. Age wasnegatively associated with gray mattervolume in network a–d and was notsignificantly associated with gray mattervolume in network e–i.
Anatomical networks in middle-aged to older adults, A. Hafkemeijer et al.1070
ª 2014 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
matter volume in five anatomical networks, predominantly containing
the temporal pole (network e), putamen (network f), and cerebellum
(networks g, h, and i), was not associated with age. The lack of age
associations in these five networks is in line with an anatomical network
study in healthy elderly (Bergfield et al., 2010). Others have shown that
the temporal areas, putamen, and cerebellum are less susceptible to age-
related differences in both gray matter volume and metabolism
(Kalpouzos et al., 2009). However, age-related differences in the
temporal anatomical network are frequently reported (Alexander et al.,
2006; Brickman et al., 2007; Montembeault et al., 2012), which makes
preservation of this network more unlikely. In this study, we found a
nonsignificant trend toward age-related gray matter volume loss in the
temporal network (network e). Further research is highly recommended
to investigate the association between age and the gray matter volume
in the temporal anatomical network.
Here, we studied whole-brain anatomical networks. The method used
in this study examines the inter-regional anatomical relationship among
spatially distributed brain structures as networks of connected regions.
This approach showed associations between age and gray matter
networks containing brain areas that were found earlier in several other
Table 1 Brain clusters of anatomical brain networks
Brain cluster†
Cluster volumeMNI coordinates
(cm3) x y z
Network a Thalamus 17.90 !2 !2 !8Cluster also contains nucleus accumbens, caudate nucleus,
hippocampus, lingual gyrus, and cerebellum
(Postcentral gyrus) 1.03 52 !8 32
(Precentral gyrus) 0.89 !20 !18 70
(Heschl’s gyrus) 0.41 !50 !26 10
Network b Lateral occipital cortex 36.76 50 !62 44
Cluster also contains precuneus and supramarginal gyrus
Cerebellum 3.17 !18 !72 !34Network c Posterior cingulate cortex 56.75 !8 22 !16
Cluster also contains paracingulate gyrus, subcallosal cortex,
operculum cortex, and precuneus
Middle temporal gyrus 6.32 56 !48 8
(Occipital fusiform gyrus) 0.42 26 !74 !14Lateral occipital cortex 0.28 !40 !72 26
Network d Anterior cingulate cortex 36.81 !2 32 28
Cluster also contains middle frontal gyrus, precentral gyrus,
and frontal medial cortex
(Cerebellum) 3.06 !20 !80 !44(Lateral occipital cortex) 2.47 50 !74 26
(Temporal pole) 1.56 !58 6 !2(Cuneus) 0.84 12 !68 24
(Precuneus) 0.68 !14 !62 22
Network e Temporal pole 29.04 !32 22 !38Cluster also contains temporal fusiform cortex
(Cerebellum) 2.59 !12 !74 !30(Anterior cingulate cortex) 1.57 10 !12 44
Network f Putamen 18.74 24 14 0
Cluster also contains caudate nucleus (and insular cortex)
Superior parietal lobule 10.40 34 !48 38
Cluster also contains lateral occipital cortex (and precuneus)
(Cerebellum) 5.37 !6 !66 !16Angular gyrus 5.35 !44 !58 20
Network g Cerebellum 24.35 42 !68 !32(Frontal pole) 0.41 52 34 !6
Network h Cerebellum 30.43 26 !64 !52(Middle frontal gyrus) 0.57 !50 28 24
(Precuneus) 0.90 20 !58 20
Network i Cerebellum 25.18 18 !86 !36Hippocampus 0.49 24 !24 !8(Postcentral gyrus) 0.49 !40 !30 40
(Frontal pole) 0.28 8 64 12
MNI, Montreal Neurological Institute 152 standard space image.
†Each gray matter anatomical network is divided into brain clusters using the cluster tool integrated in FSL. Cluster size and MNI x-, y-, and z-coordinates of each cluster are
given. Brain structures are anatomically identified using the Harvard-Oxford atlas integrated in FSL. Fig. 1 shows the most informative coronal, sagittal, and transverse slices.
Structures in parentheses in the table are not visible in Fig. 1.
Anatomical networks in middle-aged to older adults, A. Hafkemeijer et al. 1071
ª 2014 The Authors. Aging Cell published by the Anatomical Society and John Wiley & Sons Ltd.
Cerebelo
(aged 50 years and older). However, %WM demon-strated a quadratic pattern of change, increasing slightlyin subjects until the age of approximately 40 years, andit decreased thereafter. The rate of change in %WMwas significantly different in subjects younger than 50years compared with those aged 50 years or older (P !.038). Once %WM began to decline, its rate of decreasewas faster ("0.2%) than that of %GM ("0.09%). Therate of change was not significantly different for eithertotal intracranial volume or the GM/WM ratio (Fig 2C)between the two age groups.
Effects of Sex on MR Imaging VolumesSex-related differences in MR imaging volume
measurements are shown in Table 3. Significant dif-ferences were observed in total absolute intracranialvolume and absolute GM volume (P ! .003) betweenfemale subjects (1291.5 mL # 134.9 and 645.9 mL #89.6, respectively) and male subjects (1425.1 mL #138.3 and 730.6 mL # 99.3, respectively); differencesin absolute WM volume were not significant. How-ever, no significant difference was noted in either%GM or %WM between female and male subjectsafter the values were normalized for head size, al-though a trend toward slightly higher %WM in fe-male subjects (P ! .085) was present. To determinewhether the sexes differed in terms of the rate ofchange in %GM and %WM with age, results of aseparate analysis of variance was plotted (Fig 3).
Although %GM decreased with age slightly faster inmale subjects than in female subjects, the differencein the rate of change was not statistically significantfor either %GM or %WM.
DiscussionIn this study, we measured both absolute and frac-
tional %GM and %WM volumes in healthy adultsaged 20–86 years and evaluated the data by age andsex. The results showed the patterns of age-associatedchange in GM and WM volumes at a global level. Thedecline in %GM appears to occur by a relativelyyoung age (age 20 years in our study), and the de-crease is constant and linear. The %WM, in contrast,shows a quadratic pattern of change, slightly increas-
TABLE 3: Comparisons of MR imaging volume data in female andmale adults
Variables
FemaleSubjects Male Subjects
P ValuesMean SD Mean SD
Intracranial (mL) 1291.0 135.0 1425.0 138.0 .001GM (mL) 645.9 89.6 730.6 99.3 .003WM (mL) 469.8 73.3 489.1 69.6 .333% GM 50.0 4.1 51.3 5.0 .302% WM 36.4 4.0 34.4 4.1 .085GM/WM 1.4 0.2 1.5 0.3 .078
FIG 2. Regression analysis of fractional brain tissue volume esti-mates on age in 54 healthy adult subjects. Linear and weightedconstrained quadratic models are presented; these indicate theage-related volume estimates throughout adulthood in normalbrains.
A, %GM.B, %WM.C, GM/WM ratio.
1330 GE AJNR: 23, September 2002
Age-Related Total Gray Matter and WhiteMatter Changes in Normal Adult Brain.Part I: Volumetric MR Imaging Analysis
Yulin Ge, Robert I. Grossman, James S. Babb, Marcie L. Rabin,Lois J. Mannon, and Dennis L. Kolson
BACKGROUND AND PURPOSE: A technique of segmenting total gray matter (GM) and totalwhite matter (WM) in human brain is now available. We investigated the effects of age and sexon total fractional GM (%GM) and total fractional WM (%WM) volumes by using volumetricMR imaging in healthy adults.
METHODS: Fifty-four healthy volunteers (22 men, 32 women) aged 20–86 years underwentdual-echo fast spin-echo MR imaging. Total GM, total WM, and intracranial space volumeswere segmented by using MR image–based computerized semiautomated software. Volumeswere normalized as a percentage of intracranial volume (%GM and %WM) to adjust forvariations in head size. Age and sex effects were then assessed.
RESULTS: Both %GM and %WM in the intracranial space were significantly less in oldersubjects (>50 years) than in younger subjects (<50 years) (P < .0001 and P ! .02, respec-tively). Consistently, %GM decreased linearly with age, beginning in the youngest subjects.%WM decreased in a quadratic fashion, with a greater rate beginning only in adult midlife.Although larger GM volumes were observed in men before adjustments for cranium size, nosignificant differences in %GM or %WM were observed between the sexes.
CONCLUSION: GM volume loss appears to be a constant, linear function of age throughoutadult life, whereas WM volume loss seems to be delayed until middle adult life. Both appear tobe independent of sex. Quantitative analysis of %GM and %WM volumes can improve ourunderstanding of brain atrophy due to normal aging; this knowledge may be valuable indistinguishing atrophy of disease patterns from characteristics of the normal aging process.
The quantitative assessment of brain atrophy is be-coming an important consideration in monitoring theclinical outcome and treatment effects in many dis-eases, such as Alzheimer disease, multiple sclerosis,schizophrenia, alcoholism, and AIDS-related demen-tia. The reason is because recent considerable ad-vances in MR imaging and computer technology haveallowed the study of brain morphometrics in vivo,which could provide an accurate, reproducible, andquantitative measure for assessing brain atrophy.Age-associated changes in brain tissue measurements
in healthy adults have also been the subject of greatinterest in recent years, because the determination ofnormal age-specific values in brain have a role in theevaluation of both clinical-pathologic conditions andnormal aging processes. Many investigators examinethe age effects on the basis of specific brain regions,such as the corpus callosum (1), hippocampus (2),frontal and temporal lobes (3, 4), and cerebellum (5).The quantitative information from the analyses hasshown that age-related brain tissue loss may varygreatly among different brain regions (6) and be-tween the hemispheres (7). These volume measure-ments in brain tissue appear to vary with sex as well(8, 9); this observation indicates that the use of frac-tional measures to correct for population- and sex-related differences in head size is essential.
However, because brain parenchyma is generallycomposed of gray matter (GM) and white matter(WM), the quantitative analysis of brain atrophy un-derlying separate GM and WM may have implica-tions for our understanding and monitoring of theaging process in the brain. Previous groups have ex-
Received October 2, 2001; accepted after revision April 18, 2002.Supported in part by grant NS29029 from the National Institutes
of Health.From the Departments of Radiology (Y.G., R.I.G., M.L.R.,
L.J.M.) and Neurology (D.L.K), University of Pennsylvania Med-ical Center, and the Department of Biostatistics (J.S.B), Fox ChaseCancer Center, Philadelphia, PA.
Address reprints requests to Robert I. Grossman, MD, Depart-ment of Radiology, New York University School of Medicine, 550First Avenue, New York City, NY 10016.
© American Society of Neuroradiology
AJNR Am J Neuroradiol 23:1327–1333, September 2002
1327
age-related increase in shrinkage was observed in the entorhinalcortex, a region that showed no shrinkage in the younger andthe middle-age part of the sample. Moreover, hippocampal andcerebellar shrinkage increase accelerated with age, indicatingthat the oldest participants experienced particularly pro-nounced shrinkage.Age-related increases in the rate of decline are consistent
with the notion of nonlinear regional brain aging. Such increasesconfirm the cross-sectional findings that suggested an inverted-Utrajectory of lifespan change, with volume increase in youngadulthood, plateau in middle age and precipitous decline in theold age (Courchesne et al., 2000; Bartzokis et al., 2001; Jerniganet al., 2001; Jernigan and Fennema-Notestine, 2004; Raz et al.,2004b). The mid-fifties appear as a likely point of inflection ofage trend, with sizeable variation across individuals. Nonlinearshrinkage is consistent with age-related augmentation of agedifferences in other indices of white matter integrity, e.g. MRIrelaxation times (Bartzokis et al., 2003, Bartzokis 2004) or theratio of small to large myelinated axons (Tang et al., 1997). Thus,it is plausible that the regions that are late to mature and containmore thin myelinated fibers are exceedingly vulnerable to age-related declines (Raz, 2000; Bartzokis et al., 2004; Head et al.,2004). These findings underscore the importance of samplinga wide range of ages in future studies of the aging brain. Witha restricted age range, important nonlinear trends would bemissed.
The nonlinearity of brain aging trajectories may stem frommultiple factors. However, in this sample, one such factor wasidentified. In at least one region — the hippocampus — bothage-related increase and acceleration of shrinkage were limitedto participants who have received diagnosis of hypertension.Thus, if we assume that the age of onset of hypertension wasunrelated to calendar age at the time of brain assessment, theeffects of hypertension on HC volume shrinkage appearcumulative and progressive. First, the association of HC volumeshrinkage with the linear age 3 hypertension interactionsuggests that not age per se, but years spent living withhypertension, bring about the link between age and themagnitude of shrinkage. Second, the association of HC shrink-age with the quadratic age 3 hypertension interaction suggeststhat the effects of hypertension are not only cumulative but alsoprogressive. In other words, the negative effects of exposure toadditional years of hypertension increase with years of priorexposure. In the prefrontal white matter and the orbitofrontalcortex, the effects of hypertension may be more direct andadditive to the effects of age. By any account, however, theeffects of treated hypertension appear restricted to specificregions. The sources of such regional vulnerability are unclearand merit further exploration.Additional caution is in order. Hypertension results from
and is associated with multiple physiological, pathologicaland behavioral factors. Although we took care to exclude
Figure 6. Longitudinal changes in adjusted volumes of the lateral prefrontal, orbito-frontal, inferior temporal and fusiform cortices as a function of baseline age.
Cerebral Cortex November 2005, V 15 N 11 1683
at UN
AM
Direccion G
eneral de Bibliotecas on M
ay 11, 2014http://cercor.oxfordjournals.org/
Dow
nloaded from
Regional Brain Changes in Aging HealthyAdults: General Trends, IndividualDifferences and Modifiers
Naftali Raz1, Ulman Lindenberger2,3, Karen M. Rodrigue1,Kristen M. Kennedy1, Denise Head4, Adrienne Williamson5,Cheryl Dahle1, Denis Gerstorf2 and James D. Acker6
1Department of Psychology and Institute of Gerontology,Wayne State University, 87 East Ferry St, 226 Knapp Building,Detroit, MI 48202, USA 2Center for Lifespan Psychology, MaxPlanck Institute for Human Development, Berlin, Germany,3School of Psychology, Saarland University, Saarbrucken,Germany, 4Department of Psychology, Washington University,Saint-Louis, MI, USA, 5Department of Psychology, University ofMemphis, Memphis, TN, USA and 6Baptist Memorial Hospital-East, Diagnostic Imaging Center, Memphis, TN, USA
Brain aging research relies mostly on cross-sectional studies,which infer true changes from age differences. We presentlongitudinal measures of five-year change in the regional brainvolumes in healthy adults. Average and individual differences involume changes and the effects of age, sex and hypertension wereassessed with latent difference score modeling. The caudate, thecerebellum, the hippocampus and the association cortices shrunksubstantially. There was minimal change in the entorhinal and nonein the primary visual cortex. Longitudinal measures of shrinkageexceeded cross-sectional estimates. All regions except the inferiorparietal lobule showed individual differences in change. Shrinkageof the cerebellum decreased from young to middle adulthood, andincreased from middle adulthood to old age. Shrinkage of thehippocampus, the entorhinal cortices, the inferior temporal cortexand the prefrontal white matter increased with age. Moreover,shrinkage in the hippocampus and the cerebellum accelerated withage. In the hippocampus, both linear and quadratic trends inincremental age-related shrinkage were limited to the hypertensiveparticipants. Individual differences in shrinkage correlated acrosssome regions, suggesting common causes. No sex differences inage trends except for the caudate were observed. We found noevidence of neuroprotective effects of larger brain size oreducational attainment.
Keywords: cortex, hippocampus, hypertension, latent change models,white matter
Introduction
Knowledge about the aging brain is derived mostly from cross-sectional studies (Raz, 2000; Sullivan and Pfefferbaum, 2003;Hedden and Gabrieli, 2004; Raz, 2004). Such studies estimatethe average rate of aging from correlations with age but, unlikelongitudinal investigations, are incapable of directly gaugingrates of change and individual differences therein. Cross-sectional evidence suggests that in healthy adults, age-relatedvolume reduction is more pronounced in gray (especiallyprefrontal) matter, and shrinkage of sensory and entorhinalcortices is virtually nil (Raz, 2000). Thus far, longitudinal studies,with only a few exceptions, have used global indices of brainintegrity, and reveal little about regional change (Raz, 2004).Three exceptions are longitudinal studies revealing significantshrinkage of prefrontal regions with smaller but significantdeclines of other regions (Pfefferbaum et al., 1998; Resnicket al., 2003; Scahill et al., 2003). However, in those studies thesample size and/or the number of examined regions werelimited. In addition, the extant longitudinal studies, whilerelying on the samples of generally healthy adults, included
some participants with cardiovascular illness, which is commonin older persons. The effects of mild vascular conditions, whichcan exert subtle but detectable negative influence on the brainand cognition (Raz et al., 2003a), have not been examined in thecontext of longitudinal change.The studies of brain change have also been limited by reliance
on standard linear models that emphasize average trends, areoblivious to measurement issues and disregard individual differ-ences in regional changes, thus obscuring the heterogeneity inbrain aging. Those methods assume, without testing, that thesame construct is measured over time without separatingconstruct variance from specific variance and measurementerror. New longitudinal methods, such as Two-Occasion LatentDifference Modeling (LDM) (McArdle and Nesselroade, 1994),alleviate most of those problems by greatly reducing unreliabilityof difference scores, examining mean change and individualdifferences within the same framework, and formally testingmeasurement equivalence across occasions and groups (cf.Meredith, 1964). In this study, we used LDM to examine averagechanges, individual differences in change and covariances ofchange in multiple brain regions, with attention to departuresfrom linearity. We also assessed the associations of brain volumesat baseline and brain volume changes with age, sex and vascularhealth (hypertension).While some average age trends in a portionof this sample have been reported (Raz et al., 2003b,c, 2004a),most cortical regions, variability of change and the influence ofhealth-related factors have not yet been examined.On the basis of cross-sectional (Raz, 2000; Bartzokis et al.,
2001; Jernigan et al., 2001; Raz, 2004) and longitudinal(Pfefferbaum et al., 1998; Resnick et al., 2003) findings, wehypothesized the steepest decline in the lateral prefrontalcortex, with smaller shrinkage of the temporal associationcortices and sparing of the primary visual cortex and the inferiorparietal lobule. Because of reported nonlinear cross-sectionalage trends (Courchesne et al., 2000; Bartzokis et al., 2001, 2004;Jernigan et al., 2001; Raz et al., 2004b), we hypothesizedacceleration of white matter shrinkage with age. Connectivitybetween the prefrontal cortex and the striatum (Alexanderet al., 1986), and age-related shrinkage in both (Raz et al., 2003b;Rodrigue and Raz, 2004), suggested significant associationsbetween changes in those regions. In addition, we testedhypotheses that hypertension exacerbates age-related shrink-age in prefrontal regions (Raz et al., 2003a), that women showlesser brain aging thanmen (Coffey et al., 1998), that larger brainvolume is a neuroprotective factor (Satz, 1993), and that higherformal education delays brain aging and ameliorates its course(Stern et al., 1992; Kramer et al., 2004).
! The Author 2005. Published by Oxford University Press. All rights reserved.For permissions, please e-mail: [email protected]
Cerebral Cortex November 2005;15:1676--1689doi:10.1093/cercor/bhi044Advance Access publication February 9, 2005
at UN
AM
Direccion G
eneral de Bibliotecas on M
ay 11, 2014http://cercor.oxfordjournals.org/
Dow
nloaded from
© The Author 2015. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: [email protected].
1
Journals of Gerontology: MEDICAL SCIENCES, 2015, 1–7doi:10.1093/gerona/glv041
Research Article
Research Article
Midlife and Late-Life Cardiorespiratory Fitness and Brain Volume Changes in Late Adulthood: Results From the Baltimore Longitudinal Study of Aging Qu Tian,1 Stephanie A. Studenski,1 Susan M. Resnick,2 Christos Davatzikos,3 and Luigi Ferrucci1 1Longitudinal Studies Section, Translational Gerontology Branch, National Institute on Aging, Baltimore, Maryland. 2Laboratory of Behavioral Neuroscience, National Institute on Aging, Baltimore, Maryland. 3Department of Radiology, University of Pennsylvania, Philadelphia.
Address correspondence to Qu Tian, PhD, MS, 251 Bayview Blvd., Suite 100, Room 04B316, Baltimore, MD 21224. Email: [email protected]
Abstract
Background. Higher cardiorespiratory fitness (CRF) is cross-sectionally associated with more conserved brain volume in older age, but longitudinal studies are rare. This study examined whether higher midlife CRF was prospectively associated with slower atrophy, which in turn was associated with higher late-life CRF.Methods. Brain volume by magnetic resonance imaging was determined annually from 1994 to 2003 in 146 participants (Mbaseline age = 69.6 years). Peak oxygen uptake on a treadmill yielded estimated midlife CRF in 138 and late-life CRF in 73 participants.Results. Higher midlife CRF was associated with greater middle temporal gyrus, perirhinal cortex, and temporal and parietal white matter, but was not associated with atrophy progression. Slower atrophy in middle frontal and angular gyri was associated with higher late-life CRF, independent of CRF at baseline magnetic resonance imaging.Conclusions. Higher midlife CRF may play a role in preserving middle and medial temporal volumes in late adulthood. Slower atrophy in middle frontal and angular gyri may predict late-life CRF.
Key Words: Cardiovascular—Epidemiology—Neuroimaging.
Decision Editor: Stephen Kritchevsky, PhD
The intriguing relationship between increased cardiorespiratory fitness (CRF) and improved brain function, in particular executive control function and memory, is supported in human observational studies and small intervention studies (1–3). However, the mecha-nisms for brain structures involved in this relationship are not well understood, as little information is available on how brain structural changes are related to CRF.
Initial neuroimaging studies largely rely on cross-sectional designs. Current evidence shows higher CRF is associated with more conserved gray matter (GM) and white matter (WM) volumes (4–9), increased cortical thickness (9), fewer WM lesions (10), and greater WM integrity (11–14) in cognitively healthy older adults. Small intervention studies suggest aerobic exercise increases brain volumes in selected brain areas (15,16) and increases in fitness from
at University of C
alifornia, San Diego on N
ovember 15, 2015
http://biomedgerontology.oxfordjournals.org/
Dow
nloaded from
were independent of the CRF level at baseline MRI, they may suggest that such prospective associations were not accounted by prior higher fitness level. A possible cause–effect relationship may be hypothesized. These findings need to be replicated in large samples to examine whether atrophy in middle frontal and angular gyri predicts motor activity.
Greater inferior frontal and perirhinal cortex at baseline MRI and slower atrophy in superior temporal gyrus were also associated with higher late-life CRF. However, these associations were attenu-ated after further adjustment for CRF at baseline MRI. Although inferior frontal and superior temporal gyri have been previously shown to be associated with CRF in cross-sectional studies (6,15), the attenuated associations in the present study suggest that the observed associations may be explained by prior higher CRF, which is associated with higher late-life fitness level.
We did not find an association of CRF with the hippocampus. The hippocampus was identified as an important subcortical region correlated with fitness from prior cross-sectional studies (but not all) and found to be responsive to physical exercise in one small intervention study (16). Cross-sectional studies targeting hippocam-pal volume in relation to fitness in cognitively healthy older adults have yielded mixed results. Two studies observed a significant asso-ciation of fitness with hippocampal volume (5,7), while one study did not find a significant association with hippocampal volume (21).
Previous studies focused on older adults in their mid-sixties with more women than men, while the sample in our study was slightly older who were in their late sixties with more men. The null findings in the present study may also be limited by the small sample size and also by the lack of contemporary data.
We should interpret the results of the prediction of longitudi-nal regional brain volume changes from midlife CRF with caution, because there are no brain volume measures in midlife. It is possible that participants who had higher midlife CRF would have greater brain volumes in midlife, which would lead to greater brain vol-ume in late life. Future studies are needed to investigate whether prior CRF precedes brain atrophy independent of brain structure at the time of prior CRF. Second, although this study focused on brain atrophy in normal aging excluded data on and after the age of onset of dementia, the trajectories of brain structural changes in those who developed mild cognitive impairment and dementia may be different from those who did not yet experience cognitive impair-ment over the course of the study. However, it is not likely to be the case because results from the sensitivity analyses with those who did not experience MCI or dementia over the course of the study remained similar. Future studies should future examine the longitu-dinal associations between fitness and brain atrophy in large sam-ples with cognitive impairment or dementia. Because this was an
Figure 2. Scatterplots of the associations between midlife cardiorespiratory fitness and middle temporal gyrus (a), perirhinal cortex (b), parietal white matter (c), and temporal white matter (d) on average 20 y later.
Journals of Gerontology: MEDICAL SCIENCES, 2015, Vol. 00, No. 00 5
at University of C
alifornia, San Diego on N
ovember 15, 2015
http://biomedgerontology.oxfordjournals.org/
Dow
nloaded from
RESULTS
Participants
By May 2006, 58 of the original study sample (N 5 235;25% of the whole cohort) had developed cognitive impair-ment, of whom 30 (13% of the whole cohort) met ICD10criteria for probable AD. Over the same period, 31 of thesubsample (n 5 98) entered into the MRI study (32% of theimaged subsample) met ICD10 criteria for dementia of theAlzheimer type (F00), vascular dementia (F01), or atypicalor mixed Alzheimer–vascular type (F00.2). Twenty-two of56 men and 13 of 42 women had died (35% total deaths inthe imaged subsample). Sample characteristics are pre-sented according to vital status in Table 1. The results showthat brain fraction, dementia, and the ability to balance onone leg at age 78 differed significantly according to vitalstatus at age 85 years. These differences were maintainedafter adjustment for age when imaged. For survivors(n 5 63, 65% of the imaged subsample) at age 85, demen-tia was absent in 53 and present in 10 (16% of the survivingimaged subsample). Of survivors without dementia, 22were living independently in the community without mentalor physical disability (35% of the surviving imaged sub-sample). Of 35 who had died, 21 were diagnosed with de-mentia before death (60% of the deceased imagedsubsample), and 14 had died without evidence of demen-tia less than 6 months before death. Of the 31 subjects whomet criteria for dementia, 15 met criteria for probable AD;other dementia diagnoses were vascular dementia or hadimaging data to support mixed dementia pathologies (Alz-heimer’s and vascular). Table 1 shows that examination ofdifferences between survivors with and without dementialfound no differences between any of the continuous vari-ables measured (t-test Po.05) or any categorical measure(chi-square 5 188 Po.05). More women than men had de-mentia, although not significantly so (P 5.10). Those withdementia were less likely to be able to balance for 5 secondson one leg (P 5.10).
MRI and Survival
All MR images were acquired in 1999/2000. The influenceof BF (the percentage of total intracranial volume occupiedby the brain) on subsequent survival was tested using Coxproportional hazards regression using age at scanning as acovariable. This showed that BF was a significant predictorof survival (P 5.03). BF was stratified into quartiles, and thelowest quartile was compared with the three higher quar-tiles and a Kaplan-Meier graph plotted (Figure 1). Thelowest BF quartile comprised participants with BF valuesless than 0.726. The relative risk of death associated withBF less than 0.726 was 2.765 (95% confidence interval (CI)1.047–7.305) (Figure 1). The relative risks of death asso-ciated with dementia and balance were 7.96 (95%CI 5 3.06–20.68) and 3.03 (95% CI 5 1.11–8.23), respec-tively. Comparing the lowest two quartiles with the highesttwo and the highest against the other three did not yieldsignificant results.
Survival was examined using a multivariate Cox re-gression, forward conditional (Po.05) approach by firstusing dementia and sex as categorical variables and BF andage at imaging as continuous covariables. This showed thatdementia and BF were independent predictors of survival
(dementia Po.001, BF P 5.04), and that sex and age atimaging did not significantly contribute to the prediction ofdeath. Next, in addition to dementia and BF, balance wasintroduced (as a categorical yes or no variable), and it wasfound that balance was not an independent predictor ofsurvival (dementia Po.001, BF P 5.03, balance P 5.16).
MRI-Derived Regional Brain Volumes and Survival
The GM and WM regional differences between survivorsand those who died were explored using statistical para-metric mapping VBM. Using GLM and adjusting for sex,age at imaging, and TICV, the survivors were comparedwith those who died (Figure 2). Findings in gray matteridentified significant voxels (Po.05 after correction formultiple comparisons using a false detection rate (FDR)approach) bilaterally in the parietal lobes, in the left medialtemporal lobe, and in the left frontoparietal region. TheWM results also showed two significant brain regions: bi-laterally in the parietal lobes and the right posterior tem-poral region and unilaterally in the left anterior temporallobe (Po.05 FDR corrected). In each case, the uncorrectedthreshold was Po.001. The introduction of age at imaginghad no influence on the pattern of significant associations.In addition to the whole-group VBM analysis, a compar-ison was performed between the subgroups with and with-out dementia. No significant differences were foundbetween those who survived and those who died in eitherof the subgroups. This is possibly due, at least in part, to thereduced statistical power because of the small number ofsubjects.
The whole-sample VBM analysis identified areasassociated with survival. What is unclear is whether thesubsequent diagnosis of dementia could explain these as-sociations. To analyze further the significance of these re-gions, estimates of the GM and WM volumes in eachsignificant cluster were extracted for each participant fromthe segmented imaging data. The extracted values for eachcluster were first entered separately into a Cox regressionalong with sex for the whole group and the subgroups with
Age (years)84.0082.0080.0078.00
Cum
Sur
viva
l1.0
0.8
0.6
0.4
TOP
Figure 1. Kaplan-Meier graph for individuals with a brain frac-tion of 0.726 or greater (dotted line/upper three quartiles) andless than 0.726 (solid line/lowest quartile).
BRAIN VOLUME AND SURVIVAL 691JAGS APRIL 2010–VOL. 58, NO. 4
Brain Volume and Survival from Age 78 to 85: The Contribution ofAlzheimer-Type Magnetic Resonance Imaging Findings
Roger T. Staff, PhD,! Alison D. Murray,w Trevor Ahearn, PhD,z Sima Salarirad, MB,§
Donald Mowat, MB,k John M. Starr, MB,#!! Ian J. Deary, PhD,ww Helen Lemmon, MA,k andLawrence J. Whalley, MDz z
OBJECTIVES: To test the prediction of survival usingmagnetic resonance imaging (MRI)–derived global and re-gional brain volumes in subjects aged 78 to 79 withoutdementia.
DESIGN: Observational follow-up study.
SETTING: University teaching hospital.
PARTICIPANTS: Participants born in 1921, recruited in1997/98 to a longitudinal study, who underwent brain MRIin 1999/2000.
MEASUREMENTS: Vital status on May 12, 2006, globaland regional brain volumes.
RESULTS: Thirty-seven of 98 (34.9%) participants diedduring follow-up. After adjustment for cognitive ability attime of MRI examination, childhood intelligence, sex, hyper-tension, smoking history, obesity, hyperlipidemia, and age atMRI, proportion of intracranial volume occupied by thebrain (brain fraction) predicted death before age 85(P 5.04). Participants with brain fraction less than 0.726had more than twice the relative risk (2.8, 95% confidenceinterval 5 1.1–7.3) of death than participants with brainfraction greater 0.726. Lower survival was significantly as-sociated with lower gray matter volumes in bilateral parietaland left frontoparietal areas and with lower white mattervolumes in left parietal and right posterior temporal regions.Cox regression analysis showed that parietal white mattervolume (P 5.003), a subsequent diagnosis of dementia(Po.001), and sex (P 5.004) were independent predictorsof survival.
CONCLUSION: In participants aged 78 to 79, a lowerglobal brain fraction predicted survival to approximatelyage 85. Smaller regional volumetric brain reductions, seenin Alzheimer’s disease (AD), also predicted survival inde-pendent of dementia. The presence of prodromal AD prob-ably explain the main findings. J Am Geriatr Soc 58:688–695, 2010.
Key words: survival; structural MRI; dementia; riskfactors
There is a much greater contribution of dementia tomortality than is evident in conventional mortality
statistics. At age 65, the lifetime risk in both sexes of strokeor dementia is more than one in three,1 yet the exact con-tribution of Alzheimer-type neuropathology to overall mor-tality in older people remains uncertain. It is wellestablished that death certificates underreport dementia asthe underlying cause of death.2 Subclinical Alzheimer’s dis-ease (AD) makes an unknown contribution to mortality.Such subclinical disease might be indicated by volume lossof those brain regions typically affected by AD,3 but brainvolume is also likely to reflect long-standing cognitivetraits,4 so adjustment for prior cognitive ability is desirableto aid the interpretation of any mortality associations withbrain volume measures, because such long-standing cogni-tive traits themselves predict mortality.5 In addition, thereare numerous factors that can influence survival in an agedsample that need to be taken into account when estimatingthe possible contribution of magnetic resonance imaging(MRI) volumetric measures to time to death. Major influ-ences include diminished lung function6,7 and failing cog-nition.7 It is hypothesized that there is an associationbetween smaller relative volumes of those brain regionstypically affected by AD and greater mortality.
Address correspondence to Professor Lawrence J. Whalley, Institute ofApplied Health Sciences, Aberdeen, UK. E-mail: [email protected]
DOI: 10.1111/j.1532-5415.2010.02765.x
From the Departments of !Nuclear Medicine, zMagnetic Resonance Imaging,and §Radiology, Aberdeen Royal Infirmary, and zzDepartment of MentalHealth, Institute of Applied Health Sciences, University of Aberdeen,Foresterhill, Aberdeen, United Kingdom; wDepartment of Radiology,University of Aberdeen, Foresterhill, Aberdeen, United Kingdom; kRoyalCornhill Hospital, Aberdeen, United Kingdom; #Department of GeriatricMedicine, Centre for Cognitive Ageing and Cognitive Epidemiology,University of Edinburgh, Edinburgh, United Kingdom; !!Medical ResearchCouncil, Edinburgh, United Kingdom; and wwDepartment of Psychology,Centre for Cognitive Ageing and Cognitive Epidemiology, University ofEdinburgh, Edinburgh, United Kingdom.
JAGS 58:688–695, 2010r 2010, Copyright the AuthorsJournal compilation r 2010, The American Geriatrics Society 0002-8614/10/$15.00
Brain Volume and Survival from Age 78 to 85: The Contribution ofAlzheimer-Type Magnetic Resonance Imaging Findings
Roger T. Staff, PhD,! Alison D. Murray,w Trevor Ahearn, PhD,z Sima Salarirad, MB,§
Donald Mowat, MB,k John M. Starr, MB,#!! Ian J. Deary, PhD,ww Helen Lemmon, MA,k andLawrence J. Whalley, MDz z
OBJECTIVES: To test the prediction of survival usingmagnetic resonance imaging (MRI)–derived global and re-gional brain volumes in subjects aged 78 to 79 withoutdementia.
DESIGN: Observational follow-up study.
SETTING: University teaching hospital.
PARTICIPANTS: Participants born in 1921, recruited in1997/98 to a longitudinal study, who underwent brain MRIin 1999/2000.
MEASUREMENTS: Vital status on May 12, 2006, globaland regional brain volumes.
RESULTS: Thirty-seven of 98 (34.9%) participants diedduring follow-up. After adjustment for cognitive ability attime of MRI examination, childhood intelligence, sex, hyper-tension, smoking history, obesity, hyperlipidemia, and age atMRI, proportion of intracranial volume occupied by thebrain (brain fraction) predicted death before age 85(P 5.04). Participants with brain fraction less than 0.726had more than twice the relative risk (2.8, 95% confidenceinterval 5 1.1–7.3) of death than participants with brainfraction greater 0.726. Lower survival was significantly as-sociated with lower gray matter volumes in bilateral parietaland left frontoparietal areas and with lower white mattervolumes in left parietal and right posterior temporal regions.Cox regression analysis showed that parietal white mattervolume (P 5.003), a subsequent diagnosis of dementia(Po.001), and sex (P 5.004) were independent predictorsof survival.
CONCLUSION: In participants aged 78 to 79, a lowerglobal brain fraction predicted survival to approximatelyage 85. Smaller regional volumetric brain reductions, seenin Alzheimer’s disease (AD), also predicted survival inde-pendent of dementia. The presence of prodromal AD prob-ably explain the main findings. J Am Geriatr Soc 58:688–695, 2010.
Key words: survival; structural MRI; dementia; riskfactors
There is a much greater contribution of dementia tomortality than is evident in conventional mortality
statistics. At age 65, the lifetime risk in both sexes of strokeor dementia is more than one in three,1 yet the exact con-tribution of Alzheimer-type neuropathology to overall mor-tality in older people remains uncertain. It is wellestablished that death certificates underreport dementia asthe underlying cause of death.2 Subclinical Alzheimer’s dis-ease (AD) makes an unknown contribution to mortality.Such subclinical disease might be indicated by volume lossof those brain regions typically affected by AD,3 but brainvolume is also likely to reflect long-standing cognitivetraits,4 so adjustment for prior cognitive ability is desirableto aid the interpretation of any mortality associations withbrain volume measures, because such long-standing cogni-tive traits themselves predict mortality.5 In addition, thereare numerous factors that can influence survival in an agedsample that need to be taken into account when estimatingthe possible contribution of magnetic resonance imaging(MRI) volumetric measures to time to death. Major influ-ences include diminished lung function6,7 and failing cog-nition.7 It is hypothesized that there is an associationbetween smaller relative volumes of those brain regionstypically affected by AD and greater mortality.
Address correspondence to Professor Lawrence J. Whalley, Institute ofApplied Health Sciences, Aberdeen, UK. E-mail: [email protected]
DOI: 10.1111/j.1532-5415.2010.02765.x
From the Departments of !Nuclear Medicine, zMagnetic Resonance Imaging,and §Radiology, Aberdeen Royal Infirmary, and zzDepartment of MentalHealth, Institute of Applied Health Sciences, University of Aberdeen,Foresterhill, Aberdeen, United Kingdom; wDepartment of Radiology,University of Aberdeen, Foresterhill, Aberdeen, United Kingdom; kRoyalCornhill Hospital, Aberdeen, United Kingdom; #Department of GeriatricMedicine, Centre for Cognitive Ageing and Cognitive Epidemiology,University of Edinburgh, Edinburgh, United Kingdom; !!Medical ResearchCouncil, Edinburgh, United Kingdom; and wwDepartment of Psychology,Centre for Cognitive Ageing and Cognitive Epidemiology, University ofEdinburgh, Edinburgh, United Kingdom.
JAGS 58:688–695, 2010r 2010, Copyright the AuthorsJournal compilation r 2010, The American Geriatrics Society 0002-8614/10/$15.00
La corteza se adelgaza
Cerebral Cortex V 14 N 7 © Oxford University Press 2004; all rights reserved Cerebral Cortex July 2004;14:721–730; DOI: 10.1093/cercor/bhh032
Thinning of the Cerebral Cortex in Aging David H. Salat1, Randy L. Buckner2,3,5, Abraham Z. Snyder3,4, Douglas N. Greve1, Rahul S.R. Desikan1, Evelina Busa1, John C. Morris4, Anders M. Dale1 and Bruce Fischl1
1MGH/MIT/HMS Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown, MA, USA, 2Departments of Psychology, Anatomy and Neurobiology, Washington University, St Louis, MO, USA, 3Department of Radiology, Washington University, St Louis, MO, USA, 4Department of Neurology, Washington University, St Louis, MO, USA and 5Howard Hughes Medical Institute
The thickness of the cerebral cortex was measured in 106 non-demented participants ranging in age from 18 to 93 years. For eachparticipant, multiple acquisitions of structural T1-weighted magneticresonance imaging (MRI) scans were averaged to yield high-resolu-tion, high-contrast data sets. Cortical thickness was estimated as thedistance between the gray/white boundary and the outer corticalsurface, resulting in a continuous estimate across the corticalmantle. Global thinning was apparent by middle age. Men andwomen showed a similar degree of global thinning, and did not differin mean thickness in the younger or older groups. Age-associateddifferences were widespread but demonstrated a patchwork ofregional atrophy and sparing. Examination of subsets of the data fromindependent samples produced highly similar age-associatedpatterns of atrophy, suggesting that the specific anatomic patternswithin the maps were reliable. Certain results, including prominentatrophy of prefrontal cortex and relative sparing of temporal andparahippocampal cortex, converged with previous findings. Otherresults were unexpected, such as the finding of prominent atrophy infrontal cortex near primary motor cortex and calcarine cortex nearprimary visual cortex. These findings demonstrate that cortical thin-ning occurs by middle age and spans widespread cortical regionsthat include primary as well as association cortex.
Keywords: aging, atrophy, calcarine cortex, cortical thickness, dementia, executive function, magnetic resonance imaging, MRI, prefrontal cortex
IntroductionAge-related changes in brain morphology are apparent in bothpostmortem histological and in vivo magnetic resonanceimaging (MRI) studies (for reviews, see Kemper, 1994; Raz,2000). The majority of post-mortem studies report age-relatedalteration of global morphometric properties including declinein total brain weight, cortical thinning and gyral atrophy that isparticularly accelerated during the sixth and seventh decades(Kemper, 1994). Questions remain as to how early suchchanges begin and whether specific cortical regions are prefer-entially vulnerable to the morphologic changes associated withaging. In the present study, age-associated cortical atrophy wasmapped as the thinning of cortex across the entire corticalmantle, allowing visualization of regional cortical atrophypatterns.
Previous neuronal counting studies have suggested thatdegenerative changes are accelerated in specific areas of thecortex, including frontal pole and premotor cortex (Kemper,1994). Comparisons across species led to speculation that age-related cortical changes follow a gradient, with greatest and
earliest changes occurring in association areas and lesserchanges occurring later in primary sensory regions (Flood andColeman, 1988). Although early studies postulated this atrophyto be due to neurodegeneration, several recent studies suggestthat neuron number is relatively preserved in the healthy agingbrain of both humans and nonhuman primates (Morrison andHof, 1997; Peters et al., 1998), although alterations in neuronalmorphology are evident.
Contemporary in vivo neuroimaging studies have confirmedthat there are alterations in global brain morphologic proper-ties (Jernigan et al., 1991, 2001; Pfefferbaum et al., 1994;Blatter et al., 1995; Raz et al., 1997; Good et al., 2001; Sowell etal., 2003). These studies additionally support the view thatmorphological alterations may be accelerated in particularareas of the cortex — described by Raz (2000) as a ‘patchworkpattern of differential declines and relative preservation’. Pref-erential vulnerability of prefrontal cortex, in particular, hasbeen demonstrated across studies, prefrontal change beinggreater than changes in other regions (Jernigan et al., 1991; Razet al., 1997; Sowell et al., 2003). Although this preferentialvulnerability has been statistically demonstrated in certainstudies (e.g. Raz et al., 1997), the majority of MRI studies ofbrain aging have not directly compared regional effects todescribe patterns of regional selectivity.
The specific patterns of regional change place importantconstraints on what may underlie cortical atrophy and howatrophy may relate to the complex constellation of cognitivechanges associated with aging. One idea, originally proposedin the context of developmental myelination, is that age-associ-ated changes are characteristic of association cortex asopposed to primary cortex (reviewed by Kemper, 1994).Consistent with this possibility, Raz (2000) recently reported astrong relation between order of developmental myelinationand degree of age-associated volumetric atrophy, with regionsdeveloping late showing the strongest age-related atrophy.Maps of cortical atrophy, as produced in the present study,provide a test of this idea, in so far as it applies to corticalatrophy patterns. More broadly construed, maps of age-associ-ated cortical thinning provide constraints on hypothesesconcerning regionally specific processes related to atrophy.
In the present study, we measured the thickness of the cere-bral cortex from MR images (Dale and Sereno, 1993; Dale et al.,1999; Fischl et al., 1999a, 2001; Fischl and Dale, 2000), using atechnique that has been validated via histological (Rosas et al.,2002) as well as manual measurements (Kuperberg et al.,2003), to examine the regional patterns of age-associatedcortical thinning. As a secondary question, we explored the
by guest on May 11, 2014
http://cercor.oxfordjournals.org/D
ownloaded from
Salat, D. H., Buckner, R. L, Snyder, A. Z., Greve, D. N., Desikan, R. S. R., Busa, E., Morris, J. C. Dale, A. M. & Fischl, B. (2004). Thinning of the cerebral cortex in aging. Cerebral Cortex, 14, 721-730.
• Se midió el espesor de la corteza cerebral en 106 participantes sin demencia de edades comprendidas entre 18 y 93 años. • El grosor de la corteza se definió
como la distancia entre los límites de la substancia gris y blanca y la superficie cortical más externa. • El adelgazamiento cortical fue
aparente desde la edad mediana. • Los hombres y las mujeres
mostraron un grado similar de adelgazamiento global tanto en jóvenes como en viejos. • Las diferencias asociadas a la edad
fueron amplias, pero siguieron un patrón regional de atrofia. • Se incluye: áreas corticales primarias
y cortezas de asociación.
Cerebral Cortex July 2004, V 14 N 7 723
thickness was ∼0.016 mm per decade across the sampled agerange.
Gender differences in thickness and volume were exploredseparately within each of the three age groups by unpaired t-tests. These analyses revealed an effect of gender on thicknessin the left hemisphere of the MP group only t(15) = –2.22, P <0.05, and a trend in the right hemisphere (P = 0.08), with menhaving thicker cortex than women. This tentative finding ofsex differences limited to the MP group could suggest thatcortical thinning is influenced by sex hormones, as women arelikely to undergo menopause during this period and thus expe-rience a decline in hormone levels. Men generally had greatertotal cortical volume than women in both hemispheres, but nosex differences existed when values were corrected for totalintracranial volume (Fig. 2b,d).
In order to test whether change occurred early in theagespan, an ANCOVA was performed with age and gender asindependent variables while limiting the age span to partici-pants <57 years (the YP and MP groups). This analysis revealeda significant effect of age on thickness in the left and righthemispheres: F(1,44) = 4.69 and 6.45, respectively, both P <0.05. Global thickness declined with increasing age. Therewere no effects of gender or age × gender interactions forthickness in this age range. There was a significant effect of ageon volume in the left and right: F(1,44) = 9.83 and 15.58,respectively, both P < 0.01. Global cortical volume declinedwith increasing age. There were no significant effects ofgender or age × gender interactions for volume in the left andright hemispheres. Interestingly, there was a significant ageeffect on thickness in the left and right hemispheres whenlimiting the sample to participants <31 years (only YP): F(1,27)
= 8.06 and 8.02, respectively, both P = 0.01. Thus, cortical thin-ning was present as early as middle age and was apparent inthese data by the third decade of life.
Most previous neuroimaging studies have calculated totalvolumes, which can be factored into thickness and surfacearea. Thus, to further characterize the components of themorphological alterations contributing to total cortical change,we next examined age-related decline in the surface area of thecortex by ANCOVA with age and gender as independent varia-bles. There was a decline in total cortical surface area withincreasing age in the left and right hemispheres: F(1,102) =34.25 and 36.22, respectively, both P < 0.0001 (Fig. 2e). Therewas an effect of gender on surface area in the left and righthemispheres F(1,102) = 41.27 and 41.60, respectively, both P <0.0001. There were no age × gender interactions for surfacearea. When men and women were compared within eachgroup by unpaired t-test, there was a gender difference insurface area in all groups with men having greater surface areathan women in the left and right hemispheres in all groupsexamined t(29) = –2.05 and –2.01, respectively in YP, both P =0.05, t(15) = –6.73 and –6.59 respectively in MP, both P <0.001, and t(56) = –3.54 and –3.58, respectively in OP, both P< 0.001 (Fig. 2f). Thus, age-related reductions in both thicknessand surface area likely contribute to the age-related reductionsin global volume reported in prior studies. In contrast, itremains possible that developmental differences in corticalsurface area largely account for gender differences in globalbrain volumes.
Regional Measures and Maps of Cortical ThinningAge-related thinning was widespread and spanned a number ofcortical regions when thickness was regressed on age control-ling for gender (Fig. 3). Significant thinning was found inprimary sensory (occipital lobe/calcarine), primary somatosen-sory and motor (pre/post central gyrus and central sulcus) andassociation cortices (inferior lateral prefrontal cortex), withgreatest statistical significance in inferior prefrontal, precentraland supramarginal regions (Fig. 3). Thinning was qualitativelyvariable across the cortex and was regionally variable withinthe major lobes (Fig. 4a–d). Regions within the temporal lobewere relatively spared from significant thinning compared toother areas of the brain. Thickening of the cortex was alsoobserved with increasing age, although very little thickeningachieved statistical significance. These regions were mainlylocalized to the anterior cingulate and medial orbitofrontal/subcallosal cortex.
The majority of the cortical mantle showed thinning rates ofat least 0.01 mm/decade. The greatest rate (>0.07 mm/decade)was found in primary motor cortex. The greatest magnitude ofregional thinning was found in inferior prefrontal, precentral,and supramarginal regions (Fig. 5). We next tested whether thequalitative regional rates of thinning seen in the presentedmaps were statistically discernable. To do this, we createdunbiased regions of interest (ROIs) in each of the cortical lobes(frontal, parietal, occipital, and temporal). These ROIs weredefined in a subset of participants defined by splitting thepresent data set into two independent samples by ranking all ofthe participants by age (sorted by sex) and placing every otherparticipant in each group to calculate group maps. Regionsshowing maximal and minimal thinning on the effect size mapsin the first half of participants were then mapped to an inde-pendent sample of participants (the other half of the partici-
Table 1Mean global thickness measures
Measures are presented as mean, standard error of the mean and range. ***P ≤ 0.001 compared to YP (total), unpaired comparison; **P < 0.05 compared to YP (total), unpaired comparison; *P < 0.05 compared to MP (women).
Group Left thickness Right thickness
YP (men) 2.26 ± 0.020 2.24 ± 0.019
(2.12–2.35) (2.07–2.32)
YP (women) 2.26 ± 0.023 2.22 ± 0.021
(2.10–2.49) (2.05–2.48)
YP (total; n = 31) 2.26 ± 0.015 2.23 ± 0.015
(2.10–2.49) (2.05–2.48)
MP (men) 2.26* ± 0.024 2.22 ± 0.021
(2.20–2.38) (2.13–2.31)
MP (women) 2.17 ± 0.027 2.15 ± 0.025
(2.05–2.30) (2.03–2.26)
MP (total; n = 17) 2.21 ± 0.021 2.18** ± 0.019
(2.05–2.38) (2.03–2.31)
OP (men) 2.17 ± 0.021 2.14 ± 0.016
(2.03–2.28) (2.03–2.25)
OP (women) 2.16 ± 0.012 2.13 ± 0.011
(1.99–2.34) (2.01–2.31)
OP (total; n = 58) 2.16***± 0.010 2.14*** ± 0.009
(1.99–2.34) (2.01–2.31)
by guest on May 11, 2014
http://cercor.oxfordjournals.org/D
ownloaded from
Jóvenes (18-31)
Mediana edad
(41-57)
Viejos (60-93)
Disminuye la substancia blanca
BRIEF COMMUNICATION
Are Acute Infarcts the Causeof Leukoaraiosis?
Brain Mapping for 16Consecutive Weeks
John Conklin, MD, MSc,1
Frank L. Silver, MD,2
David J. Mikulis, MD,1,3 andDaniel M. Mandell, MD, PhD1,3
Neuroimaging of older adults commonly reveals abnor-mality (leukoaraiosis) in the cerebral white matter.Studies have established that extensive leukoaraiosispredicts dementia and disability, but the pathogenesisof leukoaraiosis remains unclear. We recruited 5patients with leukoaraiosis and performed magneticresonance mapping of the brain for 16 consecutiveweeks. We observed tiny lesions arising de novo in thecerebral white matter. These lesions were clinicallysilent. They had the signature features of acute ische-mic stroke. With time, the characteristics of theselesions approached those of pre-existing leukoaraiosis.Together, these findings suggest that tiny silent acuteinfarcts are a cause of leukoaraiosis.
ANN NEUROL 2014;76:899–904
Neuroimaging of older adults commonly revealspatchy or diffuse abnormality in the cerebral white
matter. Histopathologically, this represents loss of axons,myelin, and oligodendrocytes, enlargement of perivascularspaces, and glial scarring.1 Once viewed as an inconsequentialmanifestation of aging, studies have established that extensivewhite matter changes predict dementia and disability.2,3
The most widely held view is that these white mat-ter changes are caused by disease of the small cerebralarteries, with chronic hypoperfusion resulting in slowlyprogressive white matter degeneration.4,5 However, thepathogenesis is unclear and controversial.2,5–7 Manyargue that a preferable term for these changes is leukoar-aiosis, which simply means reduced white matter density,and does not denote a particular mechanism.8
Recent studies have shown that even severe carotidartery stenosis in the neck is not associated with moreextensive leukoaraiosis.9–11 This is surprising. If leukoar-aiosis is caused by chronic hypoperfusion due to diseaseof the small cerebral arteries, then coexisting carotid ste-nosis should exacerbate the blood supply inadequacy.
Whereas chronic hypoperfusion should be exacer-bated by carotid stenosis, acute occlusion of a small
cerebral artery may occur independently of carotid dis-ease. Is leukoaraiosis caused by the accumulation of tiny,asymptomatic acute infarcts? As noted in an editorial onthe topic,6 information on the detailed time scale ofdevelopment of single lesions in leukoaraiosis might pro-vide an important clue, but this intelligence is missing,despite the many population studies detailing progressionof leukoaraiosis over a period of years.
We hypothesized that serial imaging of subjectswith leukoaraiosis over a course of weeks would revealnew, tiny asymptomatic acute infarcts in the cerebralwhite matter, and that with time, these lesions wouldbecome indistinguishable from leukoaraiosis. Werecruited a small series of subjects to test this hypothesis.
Patients and Methods
This prospective observational study was approved by the insti-tutional review board of the University Health Network,Toronto. Subjects provided written informed consent.
SubjectsAttending neurologists recruited subjects from outpatient neu-rology clinics at the University Health Network over a 12-month period. Inclusion criteria were: age> 55 years, recent (6months) magnetic resonance imaging (MRI) of the brain andmagnetic resonance angiogram of the neck, and moderate tosevere leukoaraiosis (Fazekas scale! grade 2).12 Exclusion crite-ria were: cardiac source of embolism, pulmonary disease,carotid artery stenosis! 50%, and previous cortical infarct. Werecruited 5 patients who met these criteria and were willing toundergo 16 MRI examinations.
Brain MRISubjects had MRI at weekly intervals for 16 consecutive weeks.Imaging was performed on a 3.0T MRI system (Signa HDx;GE Healthcare, Little Chalfont, UK).
To map the spatial extent and severity of leukoaraiosis,we calculated maps of the transverse relaxation time (T2). T2
From the 1Department of Medical Imaging, University of Toronto;2Division of Neurology, Department of Medicine, University of Torontoand University Health Network; and 3Division of Neuroradiology,Department of Medical Imaging, University Health Network, Toronto,Ontario, Canada.
Address correspondence to Dr Mandell, Division of Neuroradiology,Toronto Western Hospital, 399 Bathurst Street, Toronto, Ontario M5T2S8, Canada. E-mail: [email protected]
Received Jun 7, 2014, and in revised form Sep 29, 2014. Accepted forpublication Sep 30, 2014.
View this article online at wileyonlinelibrary.com. DOI: 10.1002/ana.24285
VC 2014 American Neurological Association 899
https://ww
w.youtube.com
/watch?v=
J3fb0CaDpEk
reflects white matter water content and myelination, and pro-vides a quantitative and reproducible map of leukoaraiosis overserial studies. T2 maps were obtained using a multiecho fastspin-echo sequence (voxel size 5 1.72 3 1.72 3 3.00mm).
To confirm that a lesion represents acute infarction, weused diffusion tensor imaging (DTI) to measure mean diffusiv-ity (MD) and fractional anisotropy (FA), which follow an estab-lished temporal evolution in acute ischemic stroke.13–16 DTIwas performed using an echo-planar single-shot spin-echo diffu-sion-weighted imaging sequence (23 directions, voxelsize 5 1.72 3 1.72 3 3.00mm). Conventional T2-weightedfluid-attenuated inversion recovery images were also obtained.
Clinical Follow-upOne investigator, blinded to the imaging results, assessed eachsubject every fourth week to determine whether there had beenany transient or persisting deficit in strength, sensation, vision,or language.
Image AnalysisImages were imported into AFNI (NIH, http://afni.nimh.nih.gov) and spatially coregistered within each imaging session andacross all weeks. T2 maps were calculated from the multiechoT2-weighted images using a power correction method.17 FAand MD maps were calculated by least squares estimation using3D Slicer (v4.1.1, NIH, http://www.slicer.org). Segmentation ofleukoaraiosis was performed using a previously validated algo-rithm.18 The resulting regions of leukoaraiosis are consistentwith the STRIVE (STandards for ReportIng Vascular changeson nEuroimaging) definition of “whitematter hyperintensitiesof presumed vascular origin.”19 That is, lacunes, perivascularspaces, and hyperintensities within the brainstem and deep graymatter were not included in the analysis. Segmentation ofnormal-appearing white matter was performed using SPM 8(Wellcome Department of Imaging Neuroscience, UniversityCollege London, UK).
Two neuroradiologists independently reviewed alldiffusion-weighted images to identify newly appearing lesions,with discrepancies resolved by consensus. Regions of high signalintensity on diffusion-weighted images were correlated withMD maps to discriminate between true diffusion restrictionand T2 shine-through.14 All diffusion-restricting lesions weretraced, and time series data for the mean MD, FA, and T2 ofthese lesions were calculated. Lesions were categorized as arisingwithin pre-existing leukoaraiosis, arising within normal-appearing white matter, or arising at the border between leu-koaraiosis and normal-appearing white matter.
Statistical AnalysisTo analyze the temporal evolution of new lesions, time seriesdata were aligned to the week (denoted “week 0”) in whicheach new lesion was first identified. Weekly MRI parameter val-ues were grouped into the following periods: (1) week <0,before onset of a new lesion; (2) week 0, time point when anew lesion is first identified; (3) weeks 1 to 3 after lesion onset;and (4) >3 weeks after lesion onset.
Temporal changes in lesion parameters were tested using1-way repeated measures analysis of variance (ANOVA) withMD, FA, and T2 as dependent variables, and time period asthe independent variable. Mauchly’s test was used to detect sig-nificant departures from sphericity, and degrees of freedomwere corrected using Greenhouse–Geisser or Huynh–Feldtmethods as appropriate. Where the main ANOVA demon-strated a significant effect of time on a given lesion parameter,post hoc comparisons were performed between the week <0time period and each subsequent period, with Bonferroni cor-rection for multiple comparisons.
To test for localized abnormality that preceded the onsetof a newly appearing lesion, the mean MD, FA, and T2 for thelesion regions of interest in the preinjury time period werecompared to those of all normal-appearing white matter using2-tailed independent sample t tests.
ResultsThere were 3 men and 2 women. Age range was 57 to79 years. Reasons for referral to the neurology clinics
FIGURE 1: Baseline burden of leukoaraiosis, evident aspatchy regions of high signal intensity on T2-weighted(fluid-attenuated inversion recovery) magnetic resonanceimages.
ANNALS of Neurology
900 Volume 76, No. 6
Disminuye la producción de neurotransmisores
Envejecimiento y funciones dopaminérgicas
n El decremento relacionado con la edad de los marcadores de DA pre y postsinápitcos puede deberse a los siguientes factores o a una combinación de ellos:
n Decremento del número de neuronas
n Decremento del número de sinapsis por célula
n Decremento de las proteinas receptores en cada neurona
n Backman, L., y Farde, L. (2005). The role of dopamine systems in cognitive aging. En: R. Cabeza, L. Nyberg, y D. Park (Eds.). Cognitive neuroscience of aging (pp. 58-84). Nueva York: Oxford University Press.
Decremento de las neuronas dopaminérgicas
n Hay evidencia post mortem que demuestra una reducción en función de la edad, de neuronas en la substancia negra con un promedio de pérdida del 3% por década (Fearnley y Lees, 1991)
n Snow et al. (1993) hicieron un estudio con PET con F-Fluorodopa en personas con diferentes enfermedades neurodegenerativas. Luego se hizo un conteo celular post mortem en la substancia negra y encontraron una correlación entre la cantidad de neuronas y la captación del F-Fluorodopa
Decremento del número de sinapsis por célula
n El apoyo a esta hipótesis viene de la teoría que relaciona la poda sináptica con la edad (Gopnick, Meltzoff y Kuhl, 1999)
n Al nacimiento cada neurona de la corteza tiene aproximadamente 2,500 sinapsis
n A los 3 años aumenta a 15,000 n Al inicio de la adultez el número de sinapsis corticales se reduce a la mitad n Esta reducción continúa durante toda la vida adulta y en la vejez.
Decremento de las proteinas receptoras
n Estudios con roedores demuestran pérdidas importantes vinculadas con la edad en los niveles estables y síntesis de los receptores D2 mensajeros del ácido ribonucleico (mRNA
ANRV364-PS60-07 ARI 24 November 2008 18:45
The Adaptive Brain:Aging and NeurocognitiveScaffoldingDenise C. Park1 and Patricia Reuter-Lorenz2
1The Center for Brain Health, University of Texas at Dallas, Dallas, Texas 75235,email: [email protected] of Psychology, University of Michigan, Ann Arbor, Michigan 48109,email: [email protected]
Annu. Rev. Psychol. 2009. 60:173–96
The Annual Review of Psychology is online atpsych.annualreviews.org
This article’s doi:10.1146/annurev.psych.59.103006.093656
Copyright c⃝ 2009 by Annual Reviews.All rights reserved
0066-4308/09/0110-0173$20.00
Key Wordsdefault network, dedifferentiation, hippocampus, compensation,cognitive reserve, frontal activation
AbstractThere are declines with age in speed of processing, working memory,inhibitory function, and long-term memory, as well as decreases in brainstructure size and white matter integrity. In the face of these decreases,functional imaging studies have demonstrated, somewhat surprisingly,reliable increases in prefrontal activation. To account for these jointphenomena, we propose the scaffolding theory of aging and cognition(STAC). STAC provides an integrative view of the aging mind, suggest-ing that pervasive increased frontal activation with age is a marker ofan adaptive brain that engages in compensatory scaffolding in responseto the challenges posed by declining neural structures and function.Scaffolding is a normal process present across the lifespan that involvesuse and development of complementary, alternative neural circuits toachieve a particular cognitive goal. Scaffolding is protective of cogni-tive function in the aging brain, and available evidence suggests that theability to use this mechanism is strengthened by cognitive engagement,exercise, and low levels of default network engagement.
173
Ann
u. R
ev. P
sych
ol. 2
009.
60:1
73-1
96. D
ownl
oade
d fr
om w
ww
.ann
ualre
view
s.org
by U
nive
rsid
ad A
uton
oma
del E
stad
o de
Mor
elos
on
03/0
1/13
. For
per
sona
l use
onl
y.
REVIEW
How Does it STAC Up? Revisiting the Scaffolding Theoryof Aging and Cognition
Patricia A. Reuter-Lorenz & Denise C. Park
Received: 31 July 2014 /Accepted: 7 August 2014 /Published online: 21 August 2014# The Author(s) 2014. This article is published with open access at Springerlink.com
Abstract “The Scaffolding Theory of Aging and Cognition(STAC)”, proposed in 2009, is a conceptual model of cogni-tive aging that integrated evidence from structural and func-tional neuroimaging to explain how the combined effects ofadverse and compensatory neural processes produce varyinglevels of cognitive function. The model made clear and test-able predictions about how different brain variables, bothstructural and functional, were related to cognitive function,focusing on the core construct of compensatory scaffolding.The present paper provides a revised model that integratesnew evidence about the aging brain that has emerged sinceSTAC was published 5 years ago. Unlike the original STACmodel, STAC-r incorporates life-course factors that serve toenhance or deplete neural resources, thereby influencing thedevelopmental course of brain structure and function, as wellas cognition, over time. Life-course factors also influencecompensatory processes that are engaged to meet cognitivechallenge, and to ameliorate the adverse effects of structuraland functional decline. The revised model is discussed inrelation to recent lifespan and longitudinal data as well asemerging evidence about the effects of training interventions.STAC-r goes beyond the previous model by combining a life-span approach with a life-course approach to understand andpredict cognitive status and rate of cognitive change over time.
Keywords Cognitive aging . Brain imaging . Scaffolding .
Compensation
Introduction
Decades of behavioral research in the latter part of the 20thcentury characterized a variety of age-related cognitive defi-cits including memory problems, executive processing dys-function and declines in speed of processing that typify nor-mal older adults (e.g., Craik and Salthouse 2000). Despitevolumes of performance data and numerous theoretical ad-vances (e.g., Schaie et al. 1996; Schaie and Willis 2011a, b;Birren and Schaie 2005), a coherent integrated account ofcognitive aging based on behavioral data alone proved to beelusive. Fortunately, the end of the last century also broughtmajor developments in in vivo human neuroscience methods,most critically, functional and structural imaging that permit-ted scientists to relate neural activity and structural brainmeasurements to specific cognitive processing abilities(Cabeza et al. 2005). Additional and more recent advancesin imaging of white matter pathways, amyloid deposits, con-nectivity patterns, genetic, pharmacological and other bio-markers have provided a wealth of new indices of neurophys-iological status that can be integrated with behavioral perfor-mance assessments to identify the neurocognitive underpin-nings of typical age-related decline (Grady 2008; Buckneret al. 2009; Bäckman et al. 2006; Raz and Lustig 2014;Laukka et al. 2013).
In 2009 we published a model, which we referred to as theScaffolding Theory of Aging and Cognition—“STAC” forshort (Park and Reuter-Lorenz 2009). STAC aimed to explainage differences in cognitive function by incorporating theeffects of a broad range of adverse biological and neurophys-iological factors that had been associated with normal agingto date, and to delineate their dynamic interaction with pro-tective factors and newly emerging, putative compensatoryprocesses deemed to be at work in the older brain. While themodel was originally developed in the context of cross-sectional studies comparing extreme groups of younger and
P. A. Reuter-Lorenz (*)Department of Psychology, The University of Michigan, 530 ChurchStreet, Ann Arbor, MI 48109, USAe-mail: [email protected]
D. C. ParkCenter for Vital Longevity, School of Brain and Behavioral Sciences,The University of Texas at Dallas, Richardson, TX 75235, USA
Neuropsychol Rev (2014) 24:355–370DOI 10.1007/s11065-014-9270-9
Teoría del andamiaje neurocognitivo
n El andamiaje consiste en el reclutamiento de circuitos adicionales que mejoran las estructuras menguadas cuyo funcionamiento se ha vuelto ineficiente.
n Modelo comprobable y basado en la literatura científica: datos conductuales y de imágenes cerebrales.
n 7 premisas.
ANRV364-PS60-07 ARI 24 November 2008 18:45
The Adaptive Brain:Aging and NeurocognitiveScaffoldingDenise C. Park1 and Patricia Reuter-Lorenz2
1The Center for Brain Health, University of Texas at Dallas, Dallas, Texas 75235,email: [email protected] of Psychology, University of Michigan, Ann Arbor, Michigan 48109,email: [email protected]
Annu. Rev. Psychol. 2009. 60:173–96
The Annual Review of Psychology is online atpsych.annualreviews.org
This article’s doi:10.1146/annurev.psych.59.103006.093656
Copyright c⃝ 2009 by Annual Reviews.All rights reserved
0066-4308/09/0110-0173$20.00
Key Wordsdefault network, dedifferentiation, hippocampus, compensation,cognitive reserve, frontal activation
AbstractThere are declines with age in speed of processing, working memory,inhibitory function, and long-term memory, as well as decreases in brainstructure size and white matter integrity. In the face of these decreases,functional imaging studies have demonstrated, somewhat surprisingly,reliable increases in prefrontal activation. To account for these jointphenomena, we propose the scaffolding theory of aging and cognition(STAC). STAC provides an integrative view of the aging mind, suggest-ing that pervasive increased frontal activation with age is a marker ofan adaptive brain that engages in compensatory scaffolding in responseto the challenges posed by declining neural structures and function.Scaffolding is a normal process present across the lifespan that involvesuse and development of complementary, alternative neural circuits toachieve a particular cognitive goal. Scaffolding is protective of cogni-tive function in the aging brain, and available evidence suggests that theability to use this mechanism is strengthened by cognitive engagement,exercise, and low levels of default network engagement.
173
Ann
u. R
ev. P
sych
ol. 2
009.
60:1
73-1
96. D
ownl
oade
d fr
om w
ww
.ann
ualre
view
s.org
by U
nive
rsid
ad A
uton
oma
del E
stad
o de
Mor
elos
on
03/0
1/13
. For
per
sona
l use
onl
y.
Annu. Rev. Psychol., 2009, 60:173-196
47
Premisa 1
n El andamiaje es una propiedad dinámica y continua del cerebro adaptativo.
n Es una respuesta dinámica a lo largo de todo el ciclo vital a cualquier desafío, no es sólo respuesta al envejecimiento cerebral.
n Se vincula con el concepto de “reserva cognitiva” de Stern
49
Premisa 2 n La corteza prefrontal es el locus primario del
andamiaje.
n El andamiaje debe entenderse como circuitos que permiten producir una conducta o alcanzar una meta cognitiva de manera suplementaria, complementaria o a veces alternativa.
n Efecto ODFI
50
Premisa 3
n El andamiaje es una respuesta neurocognitiva a los desafíos.
n Extrínsecos (situaciones nuevas o complejas)
n Intrínsecos (cambios metabólicos)
n Transitorios (privación de sueño)
n Continuos: envejecimiento
51
Premisa 4
n Las redes del andamiaje son menos eficientes que las redes cognitivas específicas.
n Con el envejecimiento se deterioran las redes neuronales específicas y se tiene que depender más de las redes secundarias del andamiaje.
52
Premisa 5
n El cerebro envejecido es menos eficiente para generar andamiajes, la patología puede limitar por completo el establecimiento de andamiajes.
n La plasticidad es posible durante toda la vida, pero el proceso es más lento.
n Paradójicamente mientras más viejos necesitamos más andamiajes; llega un momento en el que la necesidad de andamiajes sobrepasa las capacidades de plasticidad.
53
Premisa 6
n Variabilidad inter individual y factores que promueven el andamiaje.
n El envejecimiento es multifactorial: n Susceptibilidad genética n Enfermedades (hipertensión) n Experiencias adversas n Edad avanzada
n La mayor capacidad para la construcción de andamiajes podría estar determinada por la mejor condición física, mayor estimulación cognitiva y otros factores del estilo de vida.
54
Premisa 7
n El andamiaje se promueve por el entrenamiento y la actividad cognitiva.
n Estudios con animales n Estudios con humanos
55
Evidencias sobre el uso de estrategias
56
1. ACTIVIDAD FÍSICA
¿Puede la actividad física mejorar el funcionamiento cerebral?
57
Efectos de la actividad física
58
• Aumento en las ramas dendríticas
• Desarrollo de capilares • Desarrollo de nuevas neuronas • Mejoría de la memoria y el
aprendizaje • Cambios moleculares y
neuroquímicos
5. Black JE, Isaacs KR, Anderson BJ, Alcantara AA, Greenough WT. Learning causes synaptogenesis, whereas motor activity causes angio- genesis in cerebellar cortex of adult rats. Proc Natl Acad Sci U S A. 1990;87:5568–5572. 6. Rhyu IJ, Boklewski J, Ferguson B, et al. Exercise training associated with increased cortical vascularization in adult female cynomologus monkeys. Abstr Soc Neurosci. 2003;920. 7. Cotman CW, Berchtold NC. Exercise: a behavioral intervention to enhance brain health and plasticity. Trends Neurosci. 2002;25: 295–301. 8. van Praag H, Christie BR, Sejnowski TJ, Gage FH. Running enhances neurogenesis, learning, and long-term potentiation in mice. Proc Natl Acad Sci U S A. 1999;96:13427–13431.
Efecto sobre volumen cerebral
59
n Colcombe et al. (2003). Midieron con RM los volúmenes de substancia blanca y gris en personas de 55 a 79 años y los correlacionaron con su aptitud cardiovascular.
n A mayor aptitud cardiovascular (VO2max),
menor pérdida de substancia gris en LF,
LT y LP
n Menor pérdida de substancia blanca en
tractos anteriores y posteriores
n Cambios significativos después de
correcciones estadísticas por nivel
socioeconómico, hipertensión y consumo
de tabaco, cafeína y alcohol.
60
and robust longitudinal registration approach to perform theinitial coregistration between participants’ time 1 and time 2images (17). The registration constrained spatial scaling bythe skull to minimize any potential differences in scannergeometry or misregistration due to soft-tissue changes.
First, each participant’s images were skull-stripped andsegmented into 3D maps of gray matter, white matter, andcerebrospinal fluid, using a semi-automated algorithm thattakes into account voxel intensity distributions as well ashidden Markov random fields to estimate tissue volume ateach voxel (18). Then, the 3D maps of gray and whitematters for each participant were registered to a commonspace (MNI) using a 12-parameter affine transformation.These segmented images were then used as a prioritemplates for a second-level segmentation. In addition,a mean image was calculated from all participants, spatiallysmoothed with a 12 mm full-width at half max kernel, andsubsequently used as a study-specific template. The use ofstudy-specific templates has been shown to reduce errorassociated with misregistration and, therefore, to provide abetter estimate of brain volume differences between groups.The second-level analysis then consisted of a resegmentationbased on the a priori gray and white matter maps from stage1 and a realignment to the study-specific template image.These images provide a voxel-by-voxel estimation of thevolume of gray matter, white matter, and cerebrospinal fluidcontained within the particular voxel. These images werethen multiplied by the Jacobian determinant for eachparticipant to preserve original volume and to control fordifferences in the extent of registration and possibleinterpolation error. Finally, the percent change in volumewas computed at each voxel for each participant. All ofthese processes were conducted by an experimenter whowas blind to the group assignment of each individual.
The maps representing the percent volume change in grayand white matter for each participant were then forwarded toa group analysis, where we compared the changes in volumefor aerobic exercising and nonaerobic control older adults in
a set of unpaired t tests at each voxel. We initially subjectedthe younger adult data to a simple t test against zero toevaluate whether any changes occurred during the 6-monthperiod for younger adults. These analyses yielded threestatistical parametric maps for gray and white mater, whichdescribed where (a) aerobic exercisers showed a greaterincrease in volume than stretching and toning controls, (b)nonaerobic controls showed a greater increase in volumethan aerobic exercisers, and (c) any change in volume,positive or negative, was present in younger adults. Weperformed a second set of analyses to examine whether theresults of our initial analysis interacted with the twodifferent MRI scanners used in the study. In none of theregions presented in Figure 1 did the scanner used to collectthe MRI data interact with the effects of interest. Theresulting statistical parametric maps presented in Table 2were statistically corrected for multiple comparisons at ap , .05 level for each cluster (19).
RESULTS
Descriptive information on the participants is presentedin Table 1. Participant ages ranged from 60 to 79 years, witha mean of 66.5 years. Overall, the sample was 55% female,and tended to be well educated, with an average 13.8 yearsof education. The estimated VO2 scores ranged from 12.6 to
Figure 1. Regions showing a significant increase in volume for older adults who participated in an aerobic fitness training program, compared to nonaerobic(stretching and toning) control older adults. A and B, Neurologically oriented axial slices through the brain, atþ2 andþ34 mm, respectively, in stereotaxic space. C,Sagittal slice 2 mm to the right of the midline of the brain. Blue regions: Gray matter volume was increased for aerobic exercisers, relative to nonaerobic controls.Yellow regions: White matter volume was increased for aerobic exercisers, relative to controls. (See also Table 2.)
Table 2. Cluster Size, Peak Location, and Statistical Value forEach of the Four Regions Where Aerobically Exercising
Older Adults Showed a Significant Increase in Brain Volume
Region Peak Z Cluster Size X (mm) Y (mm) Z (mm)
ACC/SMA 5.17 1459 "2 20 38
rIFG 4.01 604 54 14 30
lSTL 3.94 308 "58 "6 8
AWM 4.66 1085 4 26 2
Note: ACC/SMA ¼ anterior cingulate cortex, supplementary motor cortex;
rIFG¼ right inferior frontal gyrus; lSTL¼ left superior temporal gyrus; AWM¼anterior white matter.
1168 COLCOMBE ET AL.
at Universidad N
acional AutÃ
³noma de M
é
xico on March 31, 2013
http://biomedgerontology.oxfordjournals.org/
Dow
nloaded from
Exercise: An Active Route to Healthy Aging
Aerobic Exercise Training IncreasesBrain Volume in Aging Humans
Stanley J. Colcombe,1 Kirk I. Erickson,1 Paige E. Scalf,1 Jenny S. Kim,1 Ruchika Prakash,1
Edward McAuley,2 Steriani Elavsky,2 David X. Marquez,2 Liang Hu,2 and Arthur F. Kramer1
1Beckman Institute & Department of Psychology and 2Department of Kinesiology,University of Illinois, Urbana.
Background. The present study examined whether aerobic fitness training of older humans can increase brain volumein regions associated with age-related decline in both brain structure and cognition.
Methods. Fifty-nine healthy but sedentary community-dwelling volunteers, aged 60–79 years, participated in the 6-month randomized clinical trial. Half of the older adults served in the aerobic training group, the other half of the olderadults participated in the toning and stretching control group. Twenty young adults served as controls for the magneticresonance imaging (MRI), and did not participate in the exercise intervention. High spatial resolution estimates of grayand white matter volume, derived from 3D spoiled gradient recalled acquisition MRI images, were collected before andafter the 6-month fitness intervention. Estimates of maximal oxygen uptake (VO2) were also obtained.
Results. Significant increases in brain volume, in both gray and white matter regions, were found as a function offitness training for the older adults who participated in the aerobic fitness training but not for the older adults whoparticipated in the stretching and toning (nonaerobic) control group. As predicted, no significant changes in either gray orwhite matter volume were detected for our younger participants.
Conclusions. These results suggest that cardiovascular fitness is associated with the sparing of brain tissue in aginghumans. Furthermore, these results suggest a strong biological basis for the role of aerobic fitness in maintaining andenhancing central nervous system health and cognitive functioning in older adults.
BEGINNING in the third decade of life the human brainshows structural decline, which is disproportionately
large in the frontal, parietal, and temporal lobes of the brain(1). This decline is contemporaneously associated withdeterioration in a broad array of cognitive processes (2).Given the projected increase in the number of adults sur-viving to advanced age, and the staggering costs of caringfor older individuals who suffer from neurological decline,identifying mechanisms to offset or reverse these declineshas become increasingly important.
Cardiovascular exercise has been associated with im-proved cognitive functioning in aging humans (3,4). Theseeffects have been shown to be the greatest in higher ordercognitive processes, such as working memory, switchingbetween tasks, and inhibiting irrelevant information, all ofwhich are thought to be subserved, in part, by the frontallobes of the brain (3). However, very little is known aboutthe structural brain changes, if any, which underlie thesebenefits in humans. Previous research with nonhumananimals has shown that chronic aerobic exercise can leadto the growth of new capillaries in the brain (5,6), increasethe length and number of the dendritic interconnectionsbetween neurons (7), and even increase cell production inthe hippocampus (8). These effects likely result fromincreases in growth factors such as brain-derived neuro-trophic factor (7,9) and insulin-like growth factor (10,11),
among others (12). The end result of these structuralchanges is a better interconnected brain that is more plasticand adaptive to change (8,13). Given that cardiovascularexercise has similar effects on human cognitive function thatmight be predicted from the structural changes in nonhumananimals, it seems likely that similar structural changeswould be engendered in human brain tissue followingchronic exercise, but research examining the impact ofexercise on brain structure has overwhelmingly relied uponnonhuman animals, due to the highly invasive methodstypically required to assess changes in brain structure.
With the advent of noninvasive in vivo brain imagingtechnologies such as structural and functional magneticresonance imaging (MRI), it is possible to address questionsabout changes in the underlying brain structure of humans.In one such study (14), we found that older adults with alifelong history of cardiovascular exercise had better pre-served brains than did age-matched sedentary counterparts.Interestingly, the structural preservation was greatest in thefrontal and parietal regions of the brain, which are thoughtto subserve aspects of higher order cognition, such as workingmemory, task switching, and the inhibition of irrelevantinformation. However, owing to the cross-sectional natureof that study, it is conceivable that a number of factorsinfluence both brain volume and aerobic fitness. It is evenpossible that the relationship is reversed. That is, those older
1166
Journal of Gerontology: MEDICAL SCIENCES Copyright 2006 by The Gerontological Society of America2006, Vol. 61A, No. 11, 1166–1170
at Universidad N
acional AutÃ
³noma de M
é
xico on March 31, 2013
http://biomedgerontology.oxfordjournals.org/
Dow
nloaded from
and robust longitudinal registration approach to perform theinitial coregistration between participants’ time 1 and time 2images (17). The registration constrained spatial scaling bythe skull to minimize any potential differences in scannergeometry or misregistration due to soft-tissue changes.
First, each participant’s images were skull-stripped andsegmented into 3D maps of gray matter, white matter, andcerebrospinal fluid, using a semi-automated algorithm thattakes into account voxel intensity distributions as well ashidden Markov random fields to estimate tissue volume ateach voxel (18). Then, the 3D maps of gray and whitematters for each participant were registered to a commonspace (MNI) using a 12-parameter affine transformation.These segmented images were then used as a prioritemplates for a second-level segmentation. In addition,a mean image was calculated from all participants, spatiallysmoothed with a 12 mm full-width at half max kernel, andsubsequently used as a study-specific template. The use ofstudy-specific templates has been shown to reduce errorassociated with misregistration and, therefore, to provide abetter estimate of brain volume differences between groups.The second-level analysis then consisted of a resegmentationbased on the a priori gray and white matter maps from stage1 and a realignment to the study-specific template image.These images provide a voxel-by-voxel estimation of thevolume of gray matter, white matter, and cerebrospinal fluidcontained within the particular voxel. These images werethen multiplied by the Jacobian determinant for eachparticipant to preserve original volume and to control fordifferences in the extent of registration and possibleinterpolation error. Finally, the percent change in volumewas computed at each voxel for each participant. All ofthese processes were conducted by an experimenter whowas blind to the group assignment of each individual.
The maps representing the percent volume change in grayand white matter for each participant were then forwarded toa group analysis, where we compared the changes in volumefor aerobic exercising and nonaerobic control older adults in
a set of unpaired t tests at each voxel. We initially subjectedthe younger adult data to a simple t test against zero toevaluate whether any changes occurred during the 6-monthperiod for younger adults. These analyses yielded threestatistical parametric maps for gray and white mater, whichdescribed where (a) aerobic exercisers showed a greaterincrease in volume than stretching and toning controls, (b)nonaerobic controls showed a greater increase in volumethan aerobic exercisers, and (c) any change in volume,positive or negative, was present in younger adults. Weperformed a second set of analyses to examine whether theresults of our initial analysis interacted with the twodifferent MRI scanners used in the study. In none of theregions presented in Figure 1 did the scanner used to collectthe MRI data interact with the effects of interest. Theresulting statistical parametric maps presented in Table 2were statistically corrected for multiple comparisons at ap , .05 level for each cluster (19).
RESULTS
Descriptive information on the participants is presentedin Table 1. Participant ages ranged from 60 to 79 years, witha mean of 66.5 years. Overall, the sample was 55% female,and tended to be well educated, with an average 13.8 yearsof education. The estimated VO2 scores ranged from 12.6 to
Figure 1. Regions showing a significant increase in volume for older adults who participated in an aerobic fitness training program, compared to nonaerobic(stretching and toning) control older adults. A and B, Neurologically oriented axial slices through the brain, atþ2 andþ34 mm, respectively, in stereotaxic space. C,Sagittal slice 2 mm to the right of the midline of the brain. Blue regions: Gray matter volume was increased for aerobic exercisers, relative to nonaerobic controls.Yellow regions: White matter volume was increased for aerobic exercisers, relative to controls. (See also Table 2.)
Table 2. Cluster Size, Peak Location, and Statistical Value forEach of the Four Regions Where Aerobically Exercising
Older Adults Showed a Significant Increase in Brain Volume
Region Peak Z Cluster Size X (mm) Y (mm) Z (mm)
ACC/SMA 5.17 1459 "2 20 38
rIFG 4.01 604 54 14 30
lSTL 3.94 308 "58 "6 8
AWM 4.66 1085 4 26 2
Note: ACC/SMA ¼ anterior cingulate cortex, supplementary motor cortex;
rIFG¼ right inferior frontal gyrus; lSTL¼ left superior temporal gyrus; AWM¼anterior white matter.
1168 COLCOMBE ET AL.
at Universidad N
acional AutÃ
³noma de M
é
xico on March 31, 2013
http://biomedgerontology.oxfordjournals.org/
Dow
nloaded from
Cohorte:n=59;60y79años.Ensayoclínicoaleatorizadode6mGrupoexp:ejercicioaeróbicoGrupocontrol1:viejosanaeróbicoGrupocontrol2:jóvenessinejercMedicióndesubstblancaygrisporRM
Efectos sobre las funciones cognoscitivas
61
Meta análisis (1996-2001). 18 estudios en los que participaron personas de 55 a 80 años, sanos pero sedentarios, ejercicio supervisado, asignación aleatoria. Tamaño del efecto en: Tareas cognitivas:
• Velocidad de procesamiento (TMT, tapping) • Visoespaciales (memoria visual de Benton) • Procesos de control (go no go) • Función ejecutiva (tareas de flancos de Eriksen)
© 2003
PSYCHOLOGICAL SCIENCE
Stanley Colcombe and Arthur F. Kramer
VOL. 14, NO. 2, MARCH 2003 129
adults than younger adults in the regions of brain that are recruited tocarry out a variety of cognitive tasks (Cabeza, 2001). One suggestedexplanation for this finding is that older adults recruit additional corti-cal areas to compensate for losses in neural efficiency. Another viewcharacterizes dedifferentiation as a simple marker of cognitive de-cline. Some evidence in support of the compensatory hypothesis hasbeen provided by studies that have examined the relationship betweenperformance and brain activation. For example, Rympa and D’Esposito(2000) found, in an event-related fMRI study, that higher levels of ac-tivation of dorsolateral prefrontal cortex were associated with fasterworking memory retrieval for older adults.
Longitudinal assessments of cardiovascular changes and neurocog-nitive functioning would allow one to test the role that dedifferentia-tion plays in normal aging more directly. Such assessments wouldenable researchers to determine whether improvements in cognitivefunction that result from enhanced cardiovascular fitness would leadolder adults to become more dissimilar from younger adults in theirpatterns of brain activation (i.e., increased dedifferentiation). Alterna-tively, cardiovascular improvements might “turn back the clock,” bio-logically speaking, and lead to patterns of neural activation that aremore similar to the pattern of young adults.
The finding of significant effects for programmatic and demo-graphic moderators also provides important information concerningpotential boundary conditions on the fitness-cognition relationship,and suggests additional questions for further research. For example, itwill be important to determine whether the larger fitness benefit forolder than for younger senior citizens is the result of age differences ingeneral health or education, or is instead a function of baseline cogni-
tive and fitness levels. Similarly, the neuroprotective role of estrogen(Garcia-Segura, Cardona-Gomez, Chowen, & Azcoitia, 2000) and es-trogen replacement therapy is an important topic for further research,in light of the fact that the fitness-related cognitive benefits were largerfor women than for men. Also, the results reported here suggest thateven clinical populations of older adults can benefit cognitively fromphysical exercise. Unfortunately, the relatively small number of pub-lished clinical studies prevents closer examination of the moderatingeffects of individual physical or cognitive maladies on the efficacy ofthe training programs. Further research into this issue is clearly impor-tant and much needed. The findings regarding the moderating effectsof the type of fitness training, program duration, and training-sessionduration indicate that these factors should be systematically examinedin future intervention studies.
REFERENCES
Bangert-Drowns, R.L. (1986). Review of developments in meta-analytic method. Psycho-logical Bulletin, 99, 388–399.
*Barry, A.J., Steinmetz, J.R., Page, H.E., & Rodahl, K. (1966). The effects of physicalconditioning on older individuals: II. Motor performance and cognitive function.Journal of Gerontology, 21, 192–199.
Acknowledgments—This research was supported by grants from the Na-tional Institute on Aging (AG14966 and AG18008) and the Institute for theStudy of Aging. We would like to thank Steve Keele and an anonymous re-viewer for their helpful suggestions on our manuscript.
Fig. 1. Effect sizes for the different process-task types reflecting the four theoretical hypotheses concerning the process-based specificity of thebenefits of fitness training. Parenthetical notations on the x-axis indicate the number of effect sizes contributing to the point estimates for eachtask type in the exercise (E) and nonexercise (C) groups. Error bars show standard errors.
at FACULTAD DE FILOSOFIA Y LETRA on March 31, 2013pss.sagepub.comDownloaded from
62
Mayor en FE, mínimo en Velocidad; pero siempre E>C Mayor: Aeróbico y anaeróbico Mayor: 66-70 años
63
ume could be associated with increased levels of serum BDNF.Because the aerobic exercise group was the only group to show anincrease in volume over the 1-y period, we ran a correlation be-tween change in BDNF and change in hippocampal volume forthe aerobic exercise group to test this hypothesis. We found thatgreater changes in serum BDNF were associated with greaterincreases in volume for the left (r = 0.36; P < 0.01) and for theright (r= 0.37; P < 0.01) hippocampus (Fig. 3 C and D). Further,these effects were selective for the left (r = 0.30; P < 0.03) andright anterior hippocampus (r= 0.27; P < 0.04) and only marginalwith the left (r = 0.25; P < 0.06) and right (r = 0.22; P < 0.08)posterior hippocampus. There were no associations betweenchanges in serum BDNF and changes in caudate nucleus orthalamus volumes (all P > 0.50); nor were there any associationsbetween hippocampal volume and serumBDNF for the stretchingcontrol group (all P > 0.40). This indicates that exercise-inducedincreases in BDNF are selectively related to the changes in an-terior hippocampal volume resulting from aerobic exercise.
Hippocampal Volume Is Related to Improvements in Spatial Memory.Spatial memory (13, 22) was tested on both exercise andstretching groups at baseline, after 6 mo, and again after thecompletion of the 1-y intervention to determine whether changesin hippocampal volume translate to improved memory. Both
groups showed improvements in memory, as demonstrated bysignificant increases in accuracy between the first and last testingsessions for the aerobic exercise [t(2,51) = 2.08; P < 0.05] and thestretching control [t(2,54) = 4.41; P < 0.001] groups. Responsetimes also became faster for both groups between the baseline andpostintervention sessions (all P < 0.01), indicating that improve-ments in accuracy were not caused by changes in speed–accuracytradeoff. However, the aerobic exercise group did not improveperformance above that achieved by the stretching control group,as demonstrated by a nonsignificant Time × Group interaction[F(1,102) = 0.67; P < 0.40; ηp2 = 0.007]. Nonetheless, we foundthat higher aerobic fitness levels at baseline (r = 0.31; P < 0.001)and after intervention (r = 0.28; P < 0.004) were associated withbetter memory performance on the spatial memory task. Changein aerobic fitness levels from baseline to after intervention, how-ever, was not related to improvements in memory for either theentire sample (r= 0.15; P < 0.12) or when considering each groupseparately (both P > 0.05). Furthermore, changes in BDNF werenot associated with improvements in memory function for eithergroup (r < 0.15; P > 0.20). On the other hand, larger left and righthippocampi at baseline (both P < 0.005) and after intervention(both P < 0.005) were associated with better memory perfor-mance (12). Therefore, we reasoned that increased hippocampal
Fig. 1. (A) Example of hippocampussegmentation and graphs demonstrat-ing an increase in hippocampus volumefor the aerobic exercise group anda decrease in volume for the stretchingcontrol group. The Time × Group in-teraction was significant (P < 0.001) forboth left and right regions. (B) Exampleof caudate nucleus segmentation andgraphs demonstrating the changes involume for both groups. Although theexercise group showed an attenuationof decline, this did not reach signifi-cance (both P > 0.10). (C) Example ofthalamus segmentation and graphdemonstrating the change in volumefor both groups. None of the changeswere significant for the thalamus. Errorbars represent SEM.
Fig. 2. The exercise group showed a selective increase inthe anterior hippocampus and no change in the posteriorhippocampus. See Table 2 for Means and SDs.
Erickson et al. PNAS | February 15, 2011 | vol. 108 | no. 7 | 3019
NEU
ROSC
IENCE
Exercise training increases size of hippocampus andimproves memoryKirk I. Ericksona, Michelle W. Vossb,c, Ruchika Shaurya Prakashd, Chandramallika Basake, Amanda Szabof,Laura Chaddockb,c, Jennifer S. Kimb, Susie Heob,c, Heloisa Alvesb,c, Siobhan M. Whitef, Thomas R. Wojcickif,Emily Maileyf, Victoria J. Vieiraf, Stephen A. Martinf, Brandt D. Pencef, Jeffrey A. Woodsf, Edward McAuleyb,f,and Arthur F. Kramerb,c,1
aDepartment of Psychology, University of Pittsburgh, Pittsburgh, PA 15260; bBeckman Institute for Advanced Science and Technology, and fDepartment ofKinesiology and Community Health, University of Illinois, Champaign-Urbana, IL 61801; cDepartment of Psychology, University of Illinois, Champaign-Urbana,IL 61820; dDepartment of Psychology, Ohio State University, Columbus, OH 43210; and eDepartment of Psychology, Rice University, Houston, TX 77251
Edited* by Fred Gage, Salk Institute, San Diego, CA, and approved December 30, 2010 (received for review October 23, 2010)
The hippocampus shrinks in late adulthood, leading to impairedmemory and increased risk for dementia. Hippocampal and medialtemporal lobe volumes are larger in higher-fit adults, and physicalactivity training increases hippocampal perfusion, but the extent towhich aerobic exercise training can modify hippocampal volume inlate adulthood remains unknown. Here we show, in a randomizedcontrolled trial with 120 older adults, that aerobic exercise trainingincreases the size of the anterior hippocampus, leading to improve-ments in spatial memory. Exercise training increased hippocampalvolume by 2%, effectively reversing age-related loss in volume by1 to 2 y. We also demonstrate that increased hippocampal volumeis associated with greater serum levels of BDNF, a mediator ofneurogenesis in the dentate gyrus. Hippocampal volume declined inthe control group, but higher preintervention fitness partiallyattenuated the decline, suggesting that fitness protects againstvolume loss. Caudate nucleus and thalamus volumes were un-affected by the intervention. These theoretically important findingsindicate that aerobic exercise training is effective at reversing hip-pocampal volume loss in late adulthood, which is accompanied byimproved memory function.
aging | brain | cognition | plasticity | MRI
Deterioration of the hippocampus precedes and leads tomemory impairment in late adulthood (1, 2). Strategies to
fight hippocampal loss and protect against the development ofmemory impairment has become an important topic in recentyears from both scientific and public health perspectives. Physicalactivity, such as aerobic exercise, has emerged as a promising low-cost treatment to improve neurocognitive function that is acces-sible to most adults and is not plagued by intolerable side effectsoften found with pharmaceutical treatments (3). Exerciseenhances learning and improves retention, which is accompaniedby increased cell proliferation and survival in the hippocampus ofrodents (4–6); effects that are mediated, in part, by increasedproduction and secretion of BDNF and its receptor tyrosine ki-nase trkB (7, 8).Aerobic exercise training increases gray and white matter vol-
ume in the prefrontal cortex (9) of older adults and increases thefunctioning of key nodes in the executive control network (10, 11).Greater amounts of physical activity are associated with sparing ofprefrontal and temporal brain regions over a 9-y period, whichreduces the risk for cognitive impairment (12). Further, hippo-campal and medial temporal lobe volumes are larger in higher-fitolder adults (13, 14), and larger hippocampal volumes mediateimprovements in spatial memory (13). Exercise training increasescerebral blood volume (15) and perfusion of the hippocampus(16), but the extent to which exercise can modify the size of thehippocampus in late adulthood remains unknown.To evaluate whether exercise training increases the size of the
hippocampus and improves spatial memory, we designed a single-blind, randomized controlled trial in which adults were randomly
assigned to receive either moderate-intensity aerobic exercise 3 d/wk or stretching and toning exercises that served as a control. Wepredicted that 1 y of moderate-intensity exercise would increasethe size of the hippocampus and that change in hippocampalvolume would be associated with increased serum BDNF andimproved memory function.
ResultsAerobic Exercise Training Selectively Increases Hippocampal Volume.One hundred twenty older adults without dementia (Table 1)were randomly assigned to an aerobic exercise group (n = 60) orto a stretching control group (n = 60). Magnetic resonanceimages were collected before the intervention, after 6 mo, andagain after the completion of the program. The groups did notdiffer at baseline in hippocampal volume or attendance rates(Table 2 and SI Results). We found that the exercise interventionwas effective at increasing the size of the hippocampus. That is,the aerobic exercise group demonstrated an increase in volume ofthe left and right hippocampus by 2.12% and 1.97%, respectively,over the 1-y period, whereas the stretching control group dis-played a 1.40% and 1.43% decline over this same interval (Fig.1A). The moderating effect of aerobic exercise on hippocampalvolume loss was confirmed by a significant Time × Group in-teraction for both the left [F(2,114) = 8.25; P < 0.001; ηp2 = 0.12]and right [F(2,114) = 10.41; P < 0.001; ηp2 = 0.15] hippocampus(see Table 2 for all means and SDs).As can be seen in Fig. 2, we found that aerobic exercise selec-
tively increased the volume of the anterior hippocampus that in-cluded the dentate gyrus, where cell proliferation occurs (4, 6, 8),as well as subiculum and CA1 subfields, but had a minimal effecton the volume of the posterior section. Cells in the anterior hip-pocampus mediate acquisition of spatial memory (17) and showmore age-related atrophy compared with the tail of the hippo-campus (18, 19). The selective effect of aerobic exercise on theanterior hippocampus was confirmed by a significant Time ×Group ×Region interaction for both the left [F(2,114)= 4.05; P<0.02; ηp2 = 0.06] and right [F(2,114) = 4.67; P < 0.01; ηp2 = 0.07]hippocampus. As revealed by t tests, the aerobic exercise groupshowed an increase in anterior hippocampus volume from base-line to after intervention [left: t(2,58) = 3.38; P < 0.001; right:t(2,58) = 4.33; P < 0.001] but demonstrated no change in thevolume of the posterior hippocampus (both P > 0.10). In contrast,
Author contributions: K.I.E., M.W.V., R.S.P., C.B., J.A.W., E. McAuley, and A.F.K. designedresearch; K.I.E., M.W.V., R.S.P., A.S., L.C., J.S.K., S.H., H.A., S.M.W., T.R.W., E. Mailey, V.J.V.,S.A.M., B.D.P., E. McAuley, and A.F.K. performed research; K.I.E., M.W.V., and R.S.P. an-alyzed data; and K.I.E., M.W.V., R.S.P., and A.F.K. wrote the paper.
The authors declare no conflict of interest.
*This Direct Submission article had a prearranged editor.1To whom correspondence should be addressed. E-mail: [email protected].
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1015950108/-/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1015950108 PNAS | February 15, 2011 | vol. 108 | no. 7 | 3017–3022
NEU
ROSC
IENCE
LETTER
Failure to demonstrate that memoryimprovement is due either toaerobic exercise or increasedhippocampal volume
We read with interest the article in PNAS, “Exercise trainingincreases size of hippocampus and improves memory” byErickson et al. (1). It is a noteworthy finding that over a 1-yperiod anterior hippocampal volume increased by 2% in theaerobic exercise group, whereas it decreased by 1.4% in thestretching control group. However, contrary to both the title andabstract, there is virtually no evidence in this article that exer-cise improved memory. After 1 y there were no differences be-tween the exercise and control groups. “Both groups showedimprovements in memory, as demonstrated by significant in-creases in accuracy between the first and last testing sessions forthe aerobic exercise [t(2,51) = 2.08; P < 0.05] and the stretchingcontrol [t(2,54) = 4.41; P < 0.001] groups. Response timesalso became faster for both groups between the baseline andpostintervention sessions (all P < 0.01), indicating that im-provements in accuracy were not caused by changes in speed–accuracy tradeoff. However, the aerobic exercise group did notimprove performance above that achieved by the stretchingcontrol group, as demonstrated by a nonsignificant Time ×Group interaction [F(1,102) = 0.67; P < 0.40].” In addition,“change in aerobic fitness levels from baseline to after in-tervention was not related to improvements in memory for eitherthe entire sample (r = 0.15; P < 0.12) or when considering eachgroup separately (both P > 0.05). Furthermore, changes inBDNF were not associated with improvements in memoryfunction for either group (r < 0.15; P > 0.20).” The authors then“reasoned that increased hippocampal volume after the exercise
intervention should translate to improved memory function.”To support this hypothesis they correlated aerobic exercise grouphippocampal volume increase and memory improvement andfound statistically significant correlations, which are small. For lefthippocampus r= 0.23 (P < 0.05), accounting for 5% variance, andfor right hippocampus r = 0.29 (P < 0.02), accounting for 8%variance. They concluded that “this indicates that increases inhippocampal volume after 1 y of exercise augments memoryfunction in late adulthood.” However, association cannot be as-sumed to indicate causality. Furthermore, it is clear that althoughthe aerobic exercise group improved on the memory task, so didthe stretching control group in whom hippocampal volume de-creased, further undermining any assumed link between hippo-campal volume and improved memory. However, just such a linkwas explicitly drawn in the abstract, which states “here we show,in a randomized controlled trial with 120 older adults, that aerobicexercise training increases the size of the anterior hippocampus,leading to improvements in spatial memory.” Unfortunatelyboth the title and abstract are misleading and a major over-statement of the findings. A similar lack of precision in reporting isevident elsewhere in the research literature, making it difficult toevaluate what the real evidence is in relation to cognitive en-hancement interventions and aging. This clouds the picture for thescientific community and is misleading for the general public. Itbehooves us all to ensure rigor in our scientific reporting.
Robert F. Coena,1, Brian A. Lawlora, and RoseAnne KennybaMercer’s Institute for Research on Ageing, St. James’s Hospital,Dublin, Ireland; and bDepartment of Medical Gerontology, TrinityCollege Dublin, Ireland
1. Erickson KI, et al. (2011) Exercise training increases size of hippocampus and improvesmemory. Proc Natl Acad Sci USA 108:3017–3022.
Author contributions: R.F.C., B.A.L., and R.K. wrote the paper.
The authors declare no conflict of interest.1To whom correspondence should be addressed: E-mail: [email protected].
www.pnas.org/cgi/doi/10.1073/pnas.1102593108 PNAS | May 3, 2011 | vol. 108 | no. 18 | E89
LETTER
Failure to demonstrate that memoryimprovement is due either toaerobic exercise or increasedhippocampal volume
We read with interest the article in PNAS, “Exercise trainingincreases size of hippocampus and improves memory” byErickson et al. (1). It is a noteworthy finding that over a 1-yperiod anterior hippocampal volume increased by 2% in theaerobic exercise group, whereas it decreased by 1.4% in thestretching control group. However, contrary to both the title andabstract, there is virtually no evidence in this article that exer-cise improved memory. After 1 y there were no differences be-tween the exercise and control groups. “Both groups showedimprovements in memory, as demonstrated by significant in-creases in accuracy between the first and last testing sessions forthe aerobic exercise [t(2,51) = 2.08; P < 0.05] and the stretchingcontrol [t(2,54) = 4.41; P < 0.001] groups. Response timesalso became faster for both groups between the baseline andpostintervention sessions (all P < 0.01), indicating that im-provements in accuracy were not caused by changes in speed–accuracy tradeoff. However, the aerobic exercise group did notimprove performance above that achieved by the stretchingcontrol group, as demonstrated by a nonsignificant Time ×Group interaction [F(1,102) = 0.67; P < 0.40].” In addition,“change in aerobic fitness levels from baseline to after in-tervention was not related to improvements in memory for eitherthe entire sample (r = 0.15; P < 0.12) or when considering eachgroup separately (both P > 0.05). Furthermore, changes inBDNF were not associated with improvements in memoryfunction for either group (r < 0.15; P > 0.20).” The authors then“reasoned that increased hippocampal volume after the exercise
intervention should translate to improved memory function.”To support this hypothesis they correlated aerobic exercise grouphippocampal volume increase and memory improvement andfound statistically significant correlations, which are small. For lefthippocampus r= 0.23 (P < 0.05), accounting for 5% variance, andfor right hippocampus r = 0.29 (P < 0.02), accounting for 8%variance. They concluded that “this indicates that increases inhippocampal volume after 1 y of exercise augments memoryfunction in late adulthood.” However, association cannot be as-sumed to indicate causality. Furthermore, it is clear that althoughthe aerobic exercise group improved on the memory task, so didthe stretching control group in whom hippocampal volume de-creased, further undermining any assumed link between hippo-campal volume and improved memory. However, just such a linkwas explicitly drawn in the abstract, which states “here we show,in a randomized controlled trial with 120 older adults, that aerobicexercise training increases the size of the anterior hippocampus,leading to improvements in spatial memory.” Unfortunatelyboth the title and abstract are misleading and a major over-statement of the findings. A similar lack of precision in reporting isevident elsewhere in the research literature, making it difficult toevaluate what the real evidence is in relation to cognitive en-hancement interventions and aging. This clouds the picture for thescientific community and is misleading for the general public. Itbehooves us all to ensure rigor in our scientific reporting.
Robert F. Coena,1, Brian A. Lawlora, and RoseAnne KennybaMercer’s Institute for Research on Ageing, St. James’s Hospital,Dublin, Ireland; and bDepartment of Medical Gerontology, TrinityCollege Dublin, Ireland
1. Erickson KI, et al. (2011) Exercise training increases size of hippocampus and improvesmemory. Proc Natl Acad Sci USA 108:3017–3022.
Author contributions: R.F.C., B.A.L., and R.K. wrote the paper.
The authors declare no conflict of interest.1To whom correspondence should be addressed: E-mail: [email protected].
www.pnas.org/cgi/doi/10.1073/pnas.1102593108 PNAS | May 3, 2011 | vol. 108 | no. 18 | E89
2. ESTIMULACIÓN Y ENTRENAMIENTO COGNITIVO
¿Cuáles son los mecanismos?
64
ACTIVE (Ball et al., 2002)
n Advanced Cognitive Training for Independent and Vital Elderly
n Estudio bien controlado, n=2832 entre 65 y 94 años asignados aleatoriamente a 4 grupos n Memoria n Razonamiento n Velocidad de respuesta n Grupo control
n Seguimiento a dos y a 5 años
65
Resultados
n El entrenamiento cognitivo es efectivo y duradero n Pero restringido a las habilidades entrenadas n No demostraron tener efecto sobre las actividades
de la vida diaria n El incremento en la velocidad de procesamiento no
tuvo efecto sobre la memoria o el razonamiento n IDEM entrenamiento en MT y FE
66
Programas de computadora
n Mejoría en: atención, memoria y velocidad de respuesta.
n Se discute sobre la efectividad de estos procedimientos, aunque definitivamente son más divertidos pero:
n Mucha comercialización… n www.lumosity.es n Happy neuron n Etc. etc.
n Definitivamente es la opción de futuro
67
68
Figure 1.Benchmarking scores at baseline and after six weeks of training across the three groups ofparticipants. VSTM = Verbal short-term memory, SWM = Spatial working memory, PAL =Paired-associates learning. Bars represent standard deviations.
Owen et al. Page 9
Nature. Author manuscript; available in PMC 2010 December 10.
Europe PMC Funders A
uthor Manuscripts
Europe PMC Funders A
uthor Manuscripts
Putting brain training to the test
Adrian M. Owen1,*, Adam Hampshire1, Jessica A. Grahn1, Robert Stenton2, Said Dajani2,Alistair S. Burns3, Robert J. Howard2, and Clive G. Ballard2
1MRC Cognition and Brain Sciences Unit, 15 Chaucer Road, Cambridge, CB2 7EF, UK2King’s College London, Institute of Psychiatry, De Crespigny Park, London SE5 8AF, UK3Department of Psychiatry, University of Manchester, Manchester, M13 9PT, UK
Abstract‘Brain training’, or the quest for improved cognitive function through the regular use ofcomputerised tests, is a multimillion pound industry1, yet scientific evidence to support itsefficacy is lacking. Modest effects have been reported in some studies of older individuals2,3 andpreschool children4, and video gamers out perform non-gamers on some tests of visual attention5.However, the widely held belief that commercially available computerised brain trainers improvegeneral cognitive function in the wider population lacks empirical support. The central question isnot whether performance on cognitive tests can be improved by training, but rather, whether thosebenefits transfer to other untrained tasks or lead to any general improvement in the level ofcognitive functioning. Here we report the results of a six-week online study in which 11,430participants trained several times each week on cognitive tasks designed to improve reasoning,memory, planning, visuospatial skills and attention. Although improvements were observed inevery one of the cognitive tasks that were trained, no evidence was found for transfer effects tountrained tasks, even when those tasks were cognitively closely related.
To investigate whether regular brain training leads to any improvement in cognitivefunction, viewers of the BBC popular science programme ‘Bang Goes The Theory’participated in a six-week online study of brain training. An initial ‘benchmarking’assessment included a broad neuropsychological battery of four tests that are sensitive tochanges in cognitive function in health and disease6-12. Specifically, baseline measures ofreasoning6, verbal short-term memory (VSTM)7,12, spatial working memory (SWM)8-10,and paired-associates learning (PAL)11,13, were acquired. Participants were then randomlyassigned to one of two experimental groups or a third control group and logged on to the
*Correspondence and requests for materials should be addressed to A.M.O. ([email protected]).Author ContributionsA.M.O. co-designed the study, co-designed the training tasks, designed (with A.H.) the benchmarking tests provided bycambridgebrainscience.com, co-conducted the statistical analysis, interpreted the data and took overall responsibility for writing eachdraft of the manuscript. A.H. contributed to the design of the training tasks, designed (with A.M.O) and programmed thebenchmarking tests provided by cambridgebrainscience.com, co-conducted the statistical analysis and contributed to each draft of themanuscript. J.A.G. co-conducted the statistical analysis, contributed to the interpretation of the data, co-wrote the first draft of themanuscript and contributed to each subsequent version. RS designed the data capture, data checking and data cleaning protocols andwas responsible for converting data into a format for analysis and for the delivery of the trial database for statistical analysis. He waspart of the project management group and contributed to each draft of the manuscript. S.D. contributed significantly to the design ofthe study, piloted brain training modules, contributed significantly to the design and implementation of the recruitment and retentionstrategies, was part of the project management group and contributed to each draft of the manuscript. A.S.B. was chair of theindependent trial steering committee and advised on key aspects of study design and implementation in this capacity. He alsocontributed to each draft of the manuscript. R.J.H. advised on key aspects of general study design, contributed significantly to thedesign of the training tasks and contributed to each draft of the manuscript. C.G.B. jointly conceived of and jointly designed the studyand wrote the first draft of the protocol. He was part of the project management group, co-conducted the statistical evaluation,contributed significantly to the interpretation of the data and contributed to each draft of the manuscript.None of the authors has any competing interest.
Europe PMC Funders GroupAuthor ManuscriptNature. Author manuscript; available in PMC 2010 December 10.
Published in final edited form as:Nature. 2010 June 10; 465(7299): 775–778. doi:10.1038/nature09042.
Europe PMC Funders A
uthor Manuscripts
Europe PMC Funders A
uthor Manuscripts
n= 11,430 on line 6 semanas
69
PERSPECTIVE ARTICLEpublished: 13 September 2011doi: 10.3389/fpsyg.2011.00226
Do action video games improve perception and cognition?Walter R. Boot 1*, Daniel P. Blakely 1 and Daniel J. Simons2
1 Department of Psychology, Florida State University, Tallahassee, FL, USA2 Department of Psychology, Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL, USA
Edited by:Mattie Tops, University of Groningen,Netherlands
Reviewed by:Mattie Tops, University of Groningen,NetherlandsSarah E. Donohue, Duke University,USA
*Correspondence:Walter R. Boot, Department ofPsychology, Florida State University,1107 W. Call Street, Tallahassee, FL32306-4301, USA.e-mail: [email protected]
Frequent action video game players often outperform non-gamers on measures of per-ception and cognition, and some studies find that video game practice enhances thoseabilities. The possibility that video game training transfers broadly to other aspects of cog-nition is exciting because training on one task rarely improves performance on others. Atfirst glance, the cumulative evidence suggests a strong relationship between gaming expe-rience and other cognitive abilities, but methodological shortcomings call that conclusioninto question. We discuss these pitfalls, identify how existing studies succeed or fail inovercoming them, and provide guidelines for more definitive tests of the effects of gamingon cognition.
Keywords: video games, cognitive training, transfer of training, perceptual learning
DO ACTION VIDEO GAMES IMPROVE PERCEPTION ANDCOGNITION?Frequent action game players outperform non-gamers on a vari-ety of perceptual and cognitive measures, and some studies suggestthat video game training enhances cognitive performance on tasksother than those specific to the game (Table 1). The possibility ofbroad transfer from game training to other aspects of cognition isexciting because it countermands an extensive literature showingthat training on one task rarely improves performance on others(see Ball et al., 2002; Hertzog et al., 2009; Owen et al., 2010).
Although provocative, the conclusion that game training pro-duces unusually broad transfer is weakened by methodologicalshortcomings common to most (if not all) of the published studiesdocumenting gaming effects. The flaws we discuss are not obscureor esoteric – they are well known pitfalls in the design of clinicaltrials and experiments on expertise. Most of these shortcomingsare surmountable, but no published gaming study has successfullyavoided them all. In this perspective piece, we delineate these flawsand provide guidelines for more definitive tests of game benefits.
We focus on gaming research for three reasons: first, the claimsof broad transfer from game training diverge from typical findingsin the cognitive training literature (Hertzog et al., 2009). Second,these claims have circulated widely in the popular media and thushave had a broad impact. Third, game training holds tremendouspromise if the evidence for broad transfer of training bears out.We restrict our discussion to recent studies of the effects of actiongames on college-aged participants, but our criticisms apply tosimilar studies examining the effect of game experience on cogni-tion in children and older adults, and to studies testing the efficacyof various “brain fitness” and cognitive aging interventions.
CROSS-SECTIONAL STUDIES: COMPARING GAMERS ANDNON-GAMERSMost game training studies are premised on evidence that expertgamers outperform non-gamers on measures of perception and
cognition. Such differences are a necessary precondition for train-ing studies – if experienced gamers perform comparably to non-gamers, then there is no reason to expect game training to enhancethose abilities. Even if gamers do outperform non-gamers, the dif-ference might not be caused by gaming: people may become actiongamers because they have the types of abilities required to excelat these games, or a third factor might influence both cognitiveabilities and gaming.
One possible factor that could lead to the spurious conclusionof gaming benefits on cognition is differential expectations forexperts and novices. If gamers are recruited to a study becauseof their gaming experience, they might expect to perform wellbecause of their expertise, and a belief that you should performwell can influence performance on measures as basic as visualacuity (Langer et al., 2010). Imagine that you are recruited to par-ticipate in a study because of your gaming expertise, and the studyconsists of game like computer tasks. If you know you have beenrecruited because you are an expert, the demand characteristicsof the experimental situation will motivate you to try to performwell. In contrast, a non-gamer selected without any mention ofgaming will not experience such demand characteristics, so will beless motivated. Any difference in task performance, then, wouldbe analogous to a placebo effect.
Almost all studies comparing expert and novice gamers eitherneglect to report how subjects were recruited or make no effortto hide the nature of the study from participants. Many studiesrecruit experts through advertisements explicitly seeking peo-ple with game experience, thereby violating a core principle ofexperimental design and introducing the potential for differentialdemand characteristics (Boot et al., 2008; Colzato et al., 2010; Karleet al., 2010). The problem is amplified because gamers often arefamiliar with media and blog coverage of the benefits of gaming,so they expect to perform better when they have been recruitedfor their gaming expertise.
The danger that expectations, motivation, and prior knowl-edge drive expert/novice differences in basic task performance
www.frontiersin.org September 2011 | Volume 2 | Article 226 | 1
70
71
Training the Older Brain in 3-D: Video Game Enhances Cognitive Control UCSF Study Finds Brain Training Game Effective in Improving Multitasking Skills
72
73
74
SYSTEMS NEUROSCIENCEREVIEW ARTICLE
published: 08 April 2014doi: 10.3389/fnsys.2014.00054
Are videogame training gains specific or general?
Adam C. Oei and Michael D. Patterson*
Division of Psychology, School of Humanities and Social Sciences, Nanyang Technological University, Singapore, Singapore
Edited by:Mikhail Lebedev, Duke University, USA
Reviewed by:Richard Haier, University of California,USAKynan Eng, University of Zurich andETH Zurich, Switzerland
*Correspondence:Michael D. Patterson, Division ofPsychology, School of Humanities andSocial Sciences, NanyangTechnological University, #04-13, 14Nanyang Drive, Singapore 637332,Singaporee-mail: [email protected]
Many recent studies using healthy adults document enhancements in perception andcognition from playing commercial action videogames (AVGs). Playing action games(e.g., Call of Duty, Medal of Honor ) is associated with improved bottom-up lower-level information processing skills like visual-perceptual and attentional processes. Oneproposal states a general improvement in the ability to interpret and gather statisticalinformation to predict future actions which then leads to better performance acrossdifferent perceptual/attentional tasks. Another proposal claims all the tasks are separatelytrained in the AVGs because the AVGs and laboratory tasks contain similar demands.We review studies of action and non-AVGs to show support for the latter proposal.To explain transfer in AVGs, we argue that the perceptual and attention tasks sharecommon demands with the trained videogames (e.g., multiple object tracking (MOT),rapid attentional switches, and peripheral vision). In non-AVGs, several studies alsodemonstrate specific, limited transfer. One instance of specific transfer is the specificenhancement to mental rotation after training in games with a spatial emphasis (e.g.,Tetris). In contrast, the evidence for transfer is equivocal where the game and task donot share common demands (e.g., executive functioning). Thus, the “common demands”hypothesis of transfer not only characterizes transfer effects in AVGs, but also non-actiongames. Furthermore, such a theory provides specific predictions, which can help in theselection of games to train human cognition as well as in the design of videogamespurposed for human cognitive and perceptual enhancement. Finally this hypothesis isconsistent with the cognitive training literature where most post-training gains are fortasks similar to the training rather than general, non-specific improvements.
Keywords: video games, transfer (psychology), cognition, perception, learning
INTRODUCTIONOver the last decade, effects of videogame play on human percep-tion and cognition have been intensely studied and debated. Moststudies have examined effects from action videogame (AVG) play.With a few exceptions (e.g., Boot et al., 2008; Irons et al., 2011),results from independent laboratories have shown experiencedAVG players outperforming non-players in a variety of cognitiveand perceptual tasks (e.g., Green and Bavelier, 2003; Colzato et al.,2010; Vallett et al., 2013).
What type of games can be considered an AVG? While the com-plexity and cross-fertilization across videogames makes pigeon-holing each game into a distinct category difficult and somewhatarbitrary, AVGs contain many characteristics that make themunique. These include unpredictability, fast speed in presentationand response requirements, high perceptual load, the selectionbetween multiple action plans and an emphasis on peripheralprocessing (Green et al., 2010a; Hubert-Wallander et al., 2011).Most of the games used in AVG studies have been first-personshooters (FPS) like Call of Duty, Counterstrike, Unreal Tournamentand Medal of Honor (see also Latham et al., 2013 for more detaileddescriptions of different AVGs). Although games of other genreslike role-playing (e.g., Final Fantasy), puzzle (e.g., Tetris) mayhave one or two features in common with AVGs (e.g., speeded
responses), they rarely, if ever, present these all the aforemen-tioned demands in combination. Note that exactly what part ofthe AVG that leads to transfer is not yet clearly understood, andwhether all or only some of the components are necessary for thetransfer effects that have been observed.
Although cross-sectional comparisons suggest playingvideogames leads to cognitive enhancements, they actuallyhave little bearing on causality (Boot et al., 2011; Kristjánsson,2013). Primary problems include issues of directionality (i.e.,it is unclear whether people develop superior cognitive skillsbecause of gaming or whether people with superior skills becomegamers) and expectancy effects (people recruited for their gamingexpertise are more motivated and expect to perform better) (Bootet al., 2011; Kristjánsson, 2013).
In contrast to cross-sectional studies, longitudinal-type train-ing studies that show improved cognitive and perceptual abil-ities following a short bout of videogame training involvingnovice videogame players make stronger inferences for causality(e.g., Green and Bavelier, 2003; Wu and Spence, 2013). Thesegames used for training are so intriguing because they were notspecifically designed with the goal of training human cognitionand perception (e.g., Klingberg et al., 2005; Jaeggi et al., 2011;Anguera et al., 2013). Rather, they are commercially available
Frontiers in Systems Neuroscience www.frontiersin.org April 2014 | Volume 8 | Article 54 | 1
75
3. COMPROMISO SOCIAL
¿Influye la socialización en el mantenimiento cognitivo?
76
Compromiso y estilo de vida
n Menos sistematización de los estudios
n Correlación entre participación en actividades mental y socialmente estimulantes con la salud y la longevidad.
n Estilos de vida
n Concepciones culturales e históricas del bienestar.
77
78
• Equipos de 5-7 personas que se reunían semanalmente durante 20 semanas.
• Solución de problemas de diferentes áreas: literatura, ciencia y tecnología, ingeniería civil e historia. Ej. diseñar y construir una balsa que optimizara el peso que debe soportar.
• Evaluación neuropsicológica pre y post: velocidad de procesamiento, memoria de trabajo, razonamiento inductivo, pensamiento divergente. Medidas de compromiso social
79
80 * = P<.05
* *
*
t=3.11 p<0.001
Velocidad Mem T Razonam Hab. Espac Pens inductivo divergente
4. APRENDER NUEVAS HABILIDADES
¿Cuáles son los mecanismos?
81
82
Psychological Science2014, Vol. 25(1) 103 –112© The Author(s) 2013Reprints and permissions: sagepub.com/journalsPermissions.navDOI: 10.1177/0956797613499592pss.sagepub.com
Research Article
Despite the tremendous strides made in scientifically based recommendations for promoting physical health in adulthood, less is known about what one should do to maintain cognitive health. As baby boomers age, the issue of maintaining healthy cognitive function has become a problem of increasing social urgency. There is a considerable amount of correlational data suggesting that individuals who are engaged in intellectual and social activities in middle and late adulthood fare better cognitively than their less active peers. For example, self-reports of higher participation in cognitive, leisure, and social activities are related to better cognitive ability in middle-aged adults (Singh-Manoux, Richards, & Marmot, 2003) and are even associated with a decreased risk of being diagnosed with Alzheimer’s disease (Wilson et al., 2002; Wilson, Scherr, Schneider, Li, & Bennett, 2007).
Such results are intriguing, but there is surprisingly little evidence that lifestyle engagement maintains or
improves cognitive function (Hertzog, Kramer, Wilson, & Lindenberger, 2008). No doubt the reason is the difficulty of translating this hypothesis into an experimental design in which volunteers agree to be randomly assigned to conditions that significantly alter their daily experiences for a sustained period. Two studies to date have approached this issue. In one study, participants in the Senior Odyssey program engaged in diverse problem-solving activities in a group-based competition that spanned approximately 5 months and showed small but reliable improvements in speed of processing, inductive reasoning, and divergent thinking skills when compared with no-treatment control participants (Stine-Morrow,
499592 PSSXXX10.1177/0956797613499592Park et al.Synapse: Engagement Interventionresearch-article2013
Corresponding Author:Denise C. Park, The University of Texas at Dallas—Center for Vital Longevity, Suite 800, 1600 Viceroy Ave., Dallas, TX 75235 E-mail: [email protected]
The Impact of Sustained Engagement on Cognitive Function in Older Adults: The Synapse Project
Denise C. Park1,2, Jennifer Lodi-Smith3, Linda Drew1, Sara Haber1, Andrew Hebrank1, Gérard N. Bischof1,2, and Whitley Aamodt1
1Center for Vital Longevity, University of Texas at Dallas; 2School of Behavioral and Brain Sciences, University of Texas at Dallas; and 3Department of Psychology, Canisius College
AbstractIn the research reported here, we tested the hypothesis that sustained engagement in learning new skills that activated working memory, episodic memory, and reasoning over a period of 3 months would enhance cognitive function in older adults. In three conditions with high cognitive demands, participants learned to quilt, learned digital photography, or engaged in both activities for an average of 16.51 hr a week for 3 months. Results at posttest indicated that episodic memory was enhanced in these productive-engagement conditions relative to receptive-engagement conditions, in which participants either engaged in nonintellectual activities with a social group or performed low-demand cognitive tasks with no social contact. The findings suggest that sustained engagement in cognitively demanding, novel activities enhances memory function in older adulthood, but, somewhat surprisingly, we found limited cognitive benefits of sustained engagement in social activities.
Keywordscognitive aging, intervention, engagement, cognitive training, aging cognition, episodic memory, cognitive reserve, working memory
Received 12/18/12; Revision accepted 7/2/13
at Universidad Nacional Aut Mexic on May 10, 2014pss.sagepub.comDownloaded from
Hipótesis n La función cognitiva en personas mayores mejorará
con la dedicación intensa y sostenida (3 meses, 15 horas a la semana) al aprendizaje de nuevas habilidades en las que se activen: n La memoria de trabajo n La memoria episódica n La capacidad de razonamiento
83
Metodología
n 259 participantes 60-90 años (Media= 71.6 años). Escolaridad mínima de 10 años y mínimo de 26 en el MMSE. Sin experiencia en computación ni en las actividades que iban a aprender.
n GRUPOS EXPERIMENTALES: n Fotografía digital n Acolchamiento
n Fotografía digital / Acolchamiento
n GRUPOS DE CONTROL n Club social
n Placebo n Sin tratamiento
84
Pruebas cognitivas
n Velocidad de procesamiento. Comparación de dígitos de 3, 6 y 9 ítems.
n Control mental. Flancos de Eriksen modificado y tareas de identificación del CogState
n Memoria episódica. Test revisado de aprendizaje verbal de Hopkins, tanto para el recuerdo inmediato como demorado. Test de memoria de la batería CANTAB
n Procesamiento viso espacial. CANTAB
85
86
108 Park et al.
found that the productive-engagement groups improved more than the social group from pretest to posttest, F(1, 140) = 4.40, p < .04.
Specific effects of intervention.� The pretest and post-test transformed scores for each condition and cognitive domain are presented in Table S1. To determine the effects of different types of productive engagement, we compared each productive-engagement condition with the placebo condition. Thus, for example, for the analysis comparing the photo and placebo conditions, a 2 × 2 repeated measures ANOVA was conducted with condi-tion (photo vs. placebo) as the between-subjects variable and Time (pretest vs. posttest) as the within-subjects vari-able for each cognitive construct. In this analysis, we found a significant Condition × Time interaction for epi-sodic memory, F(1, 66) = 11.09, p = .01, with a net effect size of .54.2 We also found a marginally significant inter-action for visuospatial processing, F(1, 66) = 3.43, p = .07, with an effect size of .28, due to greater improvement in the photo condition. The analysis comparing the dual and placebo conditions also yielded a Condition × Time interaction for episodic memory, a result due to greater improvement in the dual condition, F(1, 79) = 3.83, p = .05, with a net effect size of .22. We also observed a Con-dition × Time interaction for processing speed in this analysis, F(1, 79) = 3.10, p = .05, with a net effect size of .29. No significant effects were observed in the compari-son of the quilt and placebo conditions.
Figure 2 presents gain scores for episodic memory (standardized posttest scores minus pretest scores) as a
function of condition for all of the cognitive domains. We note that when the comparisons shown in Figure 2 were corrected with a Bonferroni-Holm correction (Holm, 1979) for multiple comparisons, the only significant inter-action that remained was the episodic memory effect observed in the photo-versus-placebo comparison. We also assessed whether learning photography skills was more facilitative of cognition than socializing alone by comparing the photo condition with the social condition (rather than the placebo condition), and found that the episodic-memory effect remained significant, F(1, 63) = 8.70, p = .01.
To further explicate the intervention effect on episodic memory at the individual level, we present percent reliable change (Ball et al., 2002),3 defined as improvement on the posttest relative to the pretest that was greater than 1 stan-dard error of measurement, for each participant in the five intervention conditions (Fig. 3). Figure 3 demonstrates that the proportion of participants showing reliable improve-ment in the photo, quilt, and dual conditions was .76, .60, and .57, respectively. The social and placebo groups improved less, with the proportions of participants show-ing improvements at .47 and .46, respectively.
Discussion
The present study represents a serious attempt to change everyday lifestyles in older adults for a period of 3 months and ascertain the impact of different types of lifestyle changes on cognitive function in an elderly sample. Three of the conditions involved productive engagement, that is, participants learned novel and demanding new skills for 15 hr or more per week over the 3-month period. These conditions were contrasted with a recep-tive-engagement condition (the social control condition) in which participants engaged in novel activities and socialized for 15 hr a week but did not actively acquire new skills. This manipulation allowed us to dissociate the impact of socializing and other novel aspects of the situ-ation in the social condition from active skill and knowl-edge acquisition. This important condition has been omitted from past intervention studies that examined the impact of engagement on cognition. Additionally, the inclusion of a placebo condition, in which participants had limited social interactions and worked alone on tasks that they believed would improve cognition, provided an appropriate baseline against which to assess the impact of the other interventions.
The results can be summarized as follows. First, we found that productive engagement (in the quilt, photo, and dual conditions) caused a significant increase in epi-sodic memory compared with receptive engagement (in the social and placebo conditions). A further comparison demonstrated that the three productive-engagement
Episodic Memory
0.0
0.2
0.4
0.6
Pretest PosttestTime
Scor
e
Productive-Engagement Condition
Receptive-Engagement Condition
Fig. 1.� Normalized mean score for episodic memory as a function of condition and time. Error bars represent ±1 SE.
at Universidad Nacional Aut Mexic on May 10, 2014pss.sagepub.comDownloaded from
87
Synapse: Engagement Intervention 109
*
†
**
* †
Processing Speed Mental Control
Episodic Memory Visuospatial Processing
0.00
0.25
0.50
0.75
0.00
0.25
0.50
0.75
Photo Quilt Dual Social Placebo Photo Quilt Dual Social PlaceboCondition
Stan
dard
ized
Gain
Sco
reSt
anda
rdize
d Ga
in S
core
Condition
Fig. 2.� Mean standardized gain score as a function of condition for each cognitive construct. The standardized scores from the posttest were subtracted from standardized scores from the pretest, yielding the mean standardized gain scores for each cognitive construct. Error bars represent ±1 SE. Asterisks represent significant differences between conditions (*p = .05; **p = .01); daggers represent marginally significant differences between conditions (p = .10).
● ● ●● ● ●
● ● ● ● ● ● ●● ● ● ●
● ● ● ● ● ● ● ● ●● ●
●
●●
●●●●
●●●●●●●●●●●
●●●●●●●●●●●●●●
●●●●
●●●●●●●●●●●●●●●●●
●●●●●●●●●●●●●●●●●●●●●●●●
●
Photo Quilt Dual
Social Placebo
−1
0
1
2
−1
0
1
2
Individual Participants
Stan
dard
ized
Gain
Sco
reSt
anda
rdize
d Ga
in S
core
●
●
●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●
●●
●
●●●●●●●●●●●●●●●●●●●
●●●●●●●●●●●●●●●●
Fig. 3.� Standardized gain score for episodic memory for each participant. Results are shown separately for each condition. The dashed horizontal lines represent the standard error of measurement (the upper line is +1 SEM, and the lower line is �1 SEM). Vertical lines above the dashed horizontal line represent a reliable positive change in performance, and vertical lines below the dashed line indicate a reliable negative change in performance.
at Universidad Nacional Aut Mexic on May 10, 2014pss.sagepub.comDownloaded from
88
Synapse: Engagement Intervention 109
*
†
**
* †
Processing Speed Mental Control
Episodic Memory Visuospatial Processing
0.00
0.25
0.50
0.75
0.00
0.25
0.50
0.75
Photo Quilt Dual Social Placebo Photo Quilt Dual Social PlaceboCondition
Stan
dard
ized
Gain
Sco
reSt
anda
rdize
d Ga
in S
core
Condition
Fig. 2.� Mean standardized gain score as a function of condition for each cognitive construct. The standardized scores from the posttest were subtracted from standardized scores from the pretest, yielding the mean standardized gain scores for each cognitive construct. Error bars represent ±1 SE. Asterisks represent significant differences between conditions (*p = .05; **p = .01); daggers represent marginally significant differences between conditions (p = .10).
● ● ●● ● ●
● ● ● ● ● ● ●● ● ● ●
● ● ● ● ● ● ● ● ●● ●
●
●●
●●●●
●●●●●●●●●●●
●●●●●●●●●●●●●●
●●●●
●●●●●●●●●●●●●●●●●
●●●●●●●●●●●●●●●●●●●●●●●●
●
Photo Quilt Dual
Social Placebo
−1
0
1
2
−1
0
1
2
Individual Participants
Stan
dard
ized
Gain
Sco
reSt
anda
rdize
d Ga
in S
core
●
●
●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●
●●
●
●●●●●●●●●●●●●●●●●●●
●●●●●●●●●●●●●●●●
Fig. 3.� Standardized gain score for episodic memory for each participant. Results are shown separately for each condition. The dashed horizontal lines represent the standard error of measurement (the upper line is +1 SEM, and the lower line is �1 SEM). Vertical lines above the dashed horizontal line represent a reliable positive change in performance, and vertical lines below the dashed line indicate a reliable negative change in performance.
at Universidad Nacional Aut Mexic on May 10, 2014pss.sagepub.comDownloaded from
RESULTADOS
89
ACTIVE AGEING: A POLICY FRAMEWORK
WHO/NMH/NPH/02.8DISTR.: GENERAL
ORIG.: ENGLISH
Active AgeingA Policy Framework
World Health OrganizationNoncommunicable Diseases and Mental Health ClusterNoncommunicable Disease Prevention and Health Promotion Department Ageing and Life Course
PAGE 12
2. Active Ageing: The Concept and RationaleIf ageing is to be a positive experience, longer life must be accompanied by continu-ing opportunities for health, participation and security. The World Health Organization has adopted the term “active ageing” to express the process for achieving this vision.
What is “Active Ageing”?
Ac tive a ge ing is the p r o c e s s o f o p timizing o p p o r tunitie s fo r hea lth , p a r tic ip a tio n a nd s e c u r ity in o rde r to enha nc e qua lity o f life a s p eo p le a ge .
Active ageing applies to both individuals and population groups. It allows people to realize their potential for physical, social, and mental well being throughout the life course and to participate in society according to their needs, desires and capacities, while providing them with adequate protection, security and care when they require assistance.
The word “active” refers to continuing partici-pation in social, economic, cultural, spiritual and civic affairs, not just the ability to be physically active or to participate in the labour force. Older people who retire from work and those who are ill or live with disabilities can remain active contributors to their fami-lies, peers, communities and nations. Active
ageing aims to extend healthy life expectancy and quality of life for all people as they age, including those who are frail, disabled and in need of care.
“Health” refers to physical, mental and social well being as expressed in the WHO defi nition of health. Thus, in an active ageing frame-work, policies and programmes that promote mental health and social connections are as important as those that improve physical health status.
Maintaining autonomy and independence as one grows older is a key goal for both indi-viduals and policy makers (see box on defi ni-tions). Moreover, ageing takes place within the context of others – friends, work associ-ates, neighbours and family members. This is why interdependence as well as intergenera-tional solidarity (two-way giving and receiv-ing between individuals as well as older and younger generations) are important tenets of active ageing. Yesterday’s child is today’s adult and tomorrow’s grandmother or grandfather. The quality of life they will enjoy as grandpar-ents depends on the risks and opportunities they experienced throughout the life course, as well as the manner in which succeeding generations provide mutual aid and support when needed.
El envejecimiento activo es el proceso de optimización de las oportunidades para la salud, la participicación y la seguridad para mejorar la calidad de vida de las personas a medida que envejecen.
90
91