Australian and Newhttp://anp.sagepub.

com/
Zealand Journal of Psychiatry

The neurobiology of anorexia nervosa: A systematic review

Andrea Phillipou, Susan Lee Rossell and David Jonathan Castle
Aust N Z J Psychiatry published online 5 November 2013
DOI: 10.1177/0004867413509693

The online version of this article can be found at:

http://anp.sagepub.com/content/early/2013/10/24/0004867413509693

Published by:

http://www.sagepublications.com

On behalf of:

The Royal Australian and New Zealand College of Psychiatrists

Additional services and information for Australian and New Zealand Journal of Psychiatry can be found at:

Email Alerts: http://anp.sagepub.com/cgi/alerts

Subscriptions: http://anp.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

>> OnlineFirst Version of Record - Nov 5, 2013

What is This?

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

509693
2013
ANP00010.1177/0004867413509693ANZJP ArticlesPhillipou et al.

Review

Australian & New Zealand Journal of Psychiatry

The neurobiology of anorexia nervosa: 00(0) 1­–25

DOI: 10.1177/0004867413509693

A systematic review © The Royal Australian and

New Zealand College of Psychiatrists 2013
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
anp.sagepub.com

Andrea Phillipou1,2,3, Susan Lee Rossell4,5,6 and David Jonathan

Castle2,6

Abstract
Objective: Recent advances in neuroimaging techniques have enabled a better understanding of the neurobiological
underpinnings of anorexia nervosa (AN). The aim of this paper was to summarise our current understanding of the
neurobiology of AN.
Methods: The literature was searched using the electronic databases PubMed and Google Scholar, and by additional
hand searches through reference lists and specialist eating disorders journals. Relevant studies were included if they
were written in English, only used human participants, had a specific AN group, used clinical populations of AN, group
comparisons were reported for AN compared to healthy controls and not merely AN compared to other eating disor-
ders or other psychiatric groups, and were not case studies.
Results: The systematic review summarises a number of structural and functional brain differences which are reported
in individuals with AN, including differences in neurotransmitter function, regional cerebral blood flow, glucose metabo-
lism, volumetrics and the blood oxygen level dependent response.
Conclusion: Several structural and functional differences have been reported in AN, some of which reverse and oth-
ers which persist following weight restoration. These findings have important implications for our understanding of the
neurobiological underpinnings of AN, and further research in this field may provide new direction for the development
of more effective treatments.

Keywords
Eating disorders, neuroimaging, neurotransmitters, MRI, PET, SPECT

Introduction
Anorexia nervosa (AN) is a serious psychiatric condition under- or unresponsive. With increasing advances in tech-
with a 12-month prevalence rate of 0.4% among females, nology, particularly with the development of sophisticated
and approximately one-tenth of that among males
1Department of Optometry and Vision Sciences, The University of
(American Psychiatric Association, 2013). The crude mor-
Melbourne, Melbourne, Australia
tality rate of individuals admitted into US university hospi- 2Department of Psychiatry, The University of Melbourne, Melbourne,
tals with AN is approximately 5% per decade (American Australia
Psychiatric Association, 2013), among the highest mortal- 3Department of Mental Health, Austin Hospital, Melbourne, Australia
4Brain and Psychological Sciences Research Centre, Faculty of Life
ity rate of any psychiatric disorder (Harris and Barraclough,
1998; Sullivan, 1995). Furthermore, AN is associated with and Social Sciences, Swinburne University of Technology, Melbourne,
Australia
exceptionally high relapse rates (Eckert et al., 1995; Löwe 5Monash Alfred Psychiatry research centre (MAPrc), The Alfred
et al., 2001; Norring and Sohlberg, 1993; Strober et al., Hospital and Monash University Central Clinical School, Melbourne,
1997; Zipfel et al., 2000). A major contributing factor for Australia
6Department of Psychiatry, St Vincent’s Hospital, Melbourne, Australia
the high rates of morbidity and mortality experienced by
individuals with this condition is that the cause or causes of Corresponding author:
the illness are not clear, and although treatment modalities Andrea Phillipou, Department of Optometry and Vision Sciences, The
such as cognitive behaviour therapy and family therapy University of Melbourne, Parkville, VIC 3010, Australia.
have emerging evidence for efficacy, many patients remain Email: [email protected]

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 1 31/10/2013 10:22:17 AM

 2 ANZJP Articles

neuroimaging techniques, we are able to gain a better Structural brain differences

understanding of the neurobiological underpinnings of this
condition. Structural changes are frequently observed in the brains of
A large number of structural neuroimaging studies have individuals with AN, and are generally thought to reflect
been undertaken in AN since the mid-1980s, and with the the effects of malnutrition and starvation. The illness is
emergence of more advanced neuroimaging techniques, the associated with enlargement of the cortical sulci and ventri-
number of functional imaging studies in AN has grown rap- cles (Artmann et al., 1985; Dolan et al., 1988; Kingston
idly. Therefore, the aim of this systematic review was to et al., 1996; Swayze et al., 1996), and enlargement of the
summarise our current understanding of the neurobiology inter-hemispheric fissure (Artmann et al., 1985). Cerebral
of AN. Specifically, differences in neurotransmitter func- dystrophic changes have been found to correlate with
tion, regional cerebral blood flow, glucose metabolism, weight loss, and the reversal of these changes has also been
volumetrics and the blood oxygen level dependent (BOLD) found to correlate with the normalisation of body weight
response in AN will be reviewed. (Artmann et al., 1985; Golden et al., 1996; Kingston et al.,
1996; Swayze et al., 1996). However, one study found no
significant change in ventricular size, but a significant
Methods degree of sulcal widening after patients had attained normal
body weight (Dolan et al., 1988). A more recent study
The literature was searched using the electronic databases
found no differences in ventricular size of AN patients, but
PubMed and Google Scholar, and by additional hand
enlarged external cerebral spinal fluid (CSF) spaces when
searches through reference lists and specialist eating disor-
compared to controls (Palazidou et al., 1990). Individuals
ders journals. Journals were searched from 1980 to July
with AN have been reported to have larger total CSF vol-
2013. Search key terms included anorexia nervosa coupled
umes in the ventricles and sulci, and significantly reduced
with one or more of the following: neurobiology, neuronal,
grey and white matter volumes (Boghi et al., 2011; Castro-
neurotransmitters, serotonin, dopamine, neuropeptides,
Fornieles et al., 2009; Gaudio et al., 2011; Katzman et al.,
SPECT, PET, MRI, fMRI, functional, structural, neuroim-
1996, 1997; Lambe et al., 1997; Mainz et al., 2012; Mühlau
aging. The search resulted in an unmanageable number of
et al., 2007; Roberto et al., 2011). Differences in white mat-
hits (N > 100,000) and so only relevant studies meeting the
ter volume have not been consistently found (Castro-
eligibility criteria were included. For studies to be included
Fornieles et al., 2009; Mainz et al., 2012; Mühlau et al.,
they had to meet the following criteria: written in English;
2007). Follow-up studies in weight-recovered patients have
used only human participants; had a specific AN group and
reported elevated CSF volumes and persistent grey but not
not combined with other eating disorder patients; group
white matter deficits (Katzman et al., 1997; Lambe et al.,
comparisons were reported for AN compared to healthy
1997; Roberto et al., 2011). However, two studies have
controls, not merely AN compared to other eating disorders
reported normalisation of grey matter and CSF volume in
or psychiatric groups; used clinical populations and not non-
weight-recovered AN patients (Castro-Fornieles et al.,
clinical populations with AN symptoms; not case studies.
2009; Cowdrey et al., 2012a; Mainz et al., 2012).
Furthermore, a recent study by Lázaro et al. (2013) found
no difference in grey or white matter volume between
Results and discussion weight-recovered AN patients who were at 85% of the their
Our search strategy yielded a total of 29 and 81 publica- expected body mass index (BMI) for at least one month,
tions meeting the above eligibility criteria reporting struc- and a group of healthy controls.
tural and functional brain imaging in AN, respectively, all Specific brain structures have also been found to differ
of which are included in this review. Of the 29 studies between AN patients and controls, including reduced size
reporting structural data in AN, the three earliest studies of the pituitary gland (Doraiswamy et al., 1991), and a
utilised computerised tomography (CT), while the remain- reduction of total hippocampus-amygdala formation vol-
ing studies utilised magnetic resonance imaging (MRI). Of ume (Giordano et al., 2001). Individuals with AN have also
the 81 functional imaging studies in AN, 15 investigated been found to have decreased grey matter in the anterior
differences in neuronal systems: one utilising blood sam- cingulate cortex (ACC) (Mühlau et al., 2007), and signifi-
ples, five using lumbar puncture techniques, one utilising cantly reduced ACC volume whose degree of normalisa-
single-photon emission tomography (SPECT) and eight tion during treatment is related to outcome (McCormick
employing positron emission tomography (PET). Fifteen of et al., 2008). A more recent study has also reported a reduc-
the 81 functional imaging studies reviewed utilised SPECT, tion of grey matter volume in both individuals currently ill
seven employed PET to investigate glucose metabolism, with AN and weight-recovered patients (Friederich et al.,
and 44 of the studies reviewed utilised fMRI to study func- 2012). This study also reported reduced grey matter vol-
tional brain differences in AN. A summary of the results of umes of the amygdala and putamen in ill individuals, and
these studies are presented in Tables 1 to 4. the supplementary motor area in both weight-recovered

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 2 31/10/2013 10:22:17 AM

 Phillipou et al. 3

Table 1.  Structural brain differences in anorexia nervosa utilising magnetic resonance imaging.

Authors Participants Procedure Key findings

Amianto et al. 12 AN-R, 12 BN, All participants were AN vs HC: reduced grey matter bilaterally in the
(2013) 10 HC tested once. lateral cerebellum, precuneus, frontal orbital cortex
and cingulate cortex.

Frank et al. (2013) 19 AN-R, 24 rec- All participants were Rec-AN-R vs HC: increased insula and reduced
AN-R, 20 BN, 24 HC tested once. inferior parietal volumes.
All ED vs HC: increased gyrus rectus volume.
AN-R and rec-AN-R vs HC: reduced inferior
temporal white matter.

Lazaro et al. 35 rec-AN, 17 HC All participants were Rec-AN vs HC: no difference in grey or white
(2013) tested once. matter volume.

Yau et al. (2013) 12 rec-AN-R, 10 HC Diffusion tensor imaging Rec-AN vs HC: no significant differences in fractional
was performed. anisotropy; lower mean diffusivity in frontal, parietal
and cingulum white matter tracts.

Cowdrey et al. 16 rec-AN, 15 HC All participants were No group differences in grey matter.
(2012) tested once.

Friederich et al. 12 AN (6 AN-R, 6 All participants were AN and rec-AN vs HC: decreased grey matter
(2012) AN-BP), 13 rec-AN, tested once. volume in ACC and supplementary motor area.
14 HC AN vs HC: decreased grey matter volume in
amygdala and putamen.
AN vs rec-AN: reduced grey matter volume in the
left amygdala, putamen and bilateral inferior temporal
cortex.

Frieling et al. 21 AN (12 AN, 9 Diffusion tensor imaging AN vs HC: regional decrease in fractional anisotropy
(2012) rec-AN), 20 HC was performed. in the white matter of the posterior thalamic
radiation bilaterally and mediodorsal thalamus,
parts of the left superior longitudinal fasciculus and
bilateral posterior coronal radiate and the left middle
cerebellar peduncle.
AN vs rec-AN: no difference.

Mainz et al. (2012) T1: 19 AN (13 AN- Tested shortly after T1: higher CSF volumes, reduced cerebral grey
R, 6 AN-BP), 19 HC admission and at matter and no difference in total cerebral volume in
T2: 18 AN discharge (mean between AN.
T1 and T2 = 15 weeks). T2: grey matter volumes no longer differ.

Boghi et al. (2011) 10 AN-R first All participants were AN-R vs HC: reduced total white matter volume
presentation, 11 tested once. and focal grey matter atrophy in the hypothalamus,
AN-R more than 9 caudate nucleus, cerebellum, and frontal, parietal and
years treatment, 27 temporal areas.
HC

Gaudio et al. 16 AN-R, 16 HC All participants were AN vs HC: reduced global grey matter volume and
(2011) tested once. grey matter volume bilaterally of the middle cingulate
cortex, precuneus, and inferior and superior parietal
lobules.

Kazlouski et al. 16 AN (6 AN-BP, 10 Diffusion tensor imaging AN vs HC: reduced fractional anisotropy in the
(2011) AN-R), 17 HC was performed. bilateral fimbria-fornix, fronto-occipital fasciculus and
posterior cingulum.

Roberto et al. T1: 32 AN, 21 HC T1: within 2 weeks of Grey and white matter volume increased from T1
(2011) T2: 32 AN, 21 HC hospitalisation. to T2 in AN. Grey matter volume did not normalise
T2: before discharge. by T2.

(Continued)

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 3 31/10/2013 10:22:17 AM

 4 ANZJP Articles

Table 1. (Continued)

Authors Participants Procedure Key findings

Suchan et al. 15 AN, 15 HC All participants were AN vs HC: reduced grey matter volume of the
(2010) tested once. extrastriate body area.

Castro-Fornieles T1:12 AN, 9 HC Tested at baseline and 6 T1: higher CSF volumes in ventricles and sulci;
et al. (2009) T2: 12 AN, 9 HC month follow-up. reduced grey matter volume; no difference in white
matter volume in AN.
T2: normal CSF volumes, and grey and white matter
volumes.

McCormick et al. T1:18 AN, 18 HC Tested soon after T1: right dorsal ACC was reduced in AN.
(2008) T2: 14 AN, 18 HC admission and after T2: ACC volume normalised.
weight restoration.

Mühlau et al. 22 rec-AN, 37 HC Recovery was defined Rec-AN vs HC: global grey matter volume was
(2007) as a BMI above 17 kg/m2 decreased; no differences in white matter volume;
and regular menses for at grey matter decrease in bilateral ACC.
least 6 months.

Connan et al. 16 AN, 16 HC Tested once during AN vs HC: bilateral reduction in hippocampal
(2006) treatment. volume.

Giordano et al. 20 AN, 20 HC Tested once during AN vs HC: reduction of total hippocampus-amygdala
(2001) treatment. formation volume.

Neumärker et al. T1: 18 AN, 25 HC AN participants tested T1: significantly reduced volume of the lateral
(2000) T2: 18 AN at three different time ventricles and the fissures of the Sylvius between AN
T3: 18 AN points: at admission to and HC. The size of the mesencephalon was also
treatment (T1), with 50% markedly reduced in AN.
weight restoration (T2) T2 and 3: reduced mesencephalon size persisted.
and with normal weight
(T3). HC were tested at
one time point.

Katzman et al. T1:13 AN, 34 HC AN participants were T1: grey and white matter deficits and elevated CSF
(1997) T2: 6 AN tested at low weight (T1) volume in AN.
and 2–3 years later when T2: persistent grey matter deficit and elevated CSF
weight-recovered. HCs volume. White matter deficits no longer evident.
were tested once.

Lambe et al. 13 AN, 12 rec-AN, All participants were Rec-AN vs HC: significantly greater CSF volume and
(1997) 18 HC tested at one time point. smaller grey matter volume.
Weight recovery ranged Rec-AN vs AN: smaller CSF volumes and larger grey
from 1 to 23 years. and white matter volumes.

Golden et al. T1: 12 AN, 12 HC AN participants T1: reduced total ventricular volume was decreased
(1996) T2: 12 AN were tested during in AN.
hospitalisation and T2: following re-feeding, ventricular volume returned
following nutritional to the normal range.
rehabilitation. HC were
tested once.

Katzman et al. 13 AN, 8 HC All participants were AN vs HC: larger total CSF volumes, and reduced
(1996) tested at one time point. total grey and white matter volume.

Kingston et al. T1: 36 AN, 41 HC AN participants were T1: AN had enlarged lateral ventricles and dilated
(1996) T2: 33 rec-AN, 41 tested while underweight sulci.
HC and after having gained at T2: enlarged ventricles and dilated sulci in AN
least 10% of their body returned to normal levels.
weight. HC controls were
tested on two occasions
with a similar time interval.

(Continued)

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 4 31/10/2013 10:22:17 AM

 Phillipou et al. 5

Table 1. (Continued)

Authors Participants Procedure Key findings

Swayze et al. T1: 10 AN, 10 HC AN participants were T1: AN had significantly larger ventricles at and did
(1996) T2: 10 rec-AN tested within a mean not differ in total brain volume from HC.
of 3.9 days following T2: ventricular volume decreased and brain volume
admission and at weight increased significantly.
recovery (BMI of at least
18). HC were tested
once.

Doraiswamy et al. 14 AN, 12 BN, 14 All participants were AN and BN vs HC: reduced pituitary gland size.
(1991) HC tested once.

AN: anorexia nervosa; BN: bulimia nervosa; rec-AN: weight-recovered AN; HC: healthy controls; CSF: cerebrospinal fluid; BMI: body mass index;
T1: time 1; T2: time 2; T3: time 3; ACC: anterior cingulate cortex.

Table 2.  Structural brain differences in anorexia nervosa utilising computed axial tomography.

Authors Participants Procedure Key findings

Palazidou et al. (1990) 17 AN, 9 HC All participants were tested once. AN vs HC: enlarged external CSF
spaces; ventricular size did not differ.

Dolan et al. (1988) T1: 25 AN, 17 HC AN participants were tested while T1: AN had enlarged ventricles and
T2: 14 rec-AN underweight and after weight sulci widening.
recovery. T2: Sulcal widening lessened after
weight recovery but ventricular
appearance did not significantly change.

Artmann et al. (1985) T1: 35 AN AN participants were tested during T1: enlargements of CSF spaces,
T2: 26 AN hospital admission (T1), following 3 sulci and ventricles, and expansion of
T3: 3 AN weeks to 10 months (T2) and 2–3 interhemispheric and Sylvian fissures
months after first examination for in AN.
participants with early T2 times. T2 and T3: changes in CSF spaces,
sulci and ventricles reverse.

AN: anorexia nervosa; rec-AN: weight-recovered AN; HC: healthy controls; CSF: cerebrospinal fluid; T1: time 1; T2: time 2; T3: time 3.

Table 3.  Functional brain differences in AN utilising functional magnetic resonance imaging.

Authors Participants Procedure Key findings

Amianto et al. 12 AN-R, 12 BN, Resting state scans were performed. AN vs HC: intrinsic connectivity networks
(2013) 10 HC in the cerebellum were more restricted
to vermian areas; greater medial and less
lateral connectivity in the cerebellum;
more connectivity with the bilateral
temporal pole, posterior cingulate cortex
and left insula, and less connectivity with
the parietal lobe.

Fonville et al. 35 AN (28 AN-R, Participants completed the AN vs HC: poorer on the simple and
(2013) 7 AN-BP), 37 HC embedded figures test where a target complex trials and had greater activation
geometrical shape and two more of the fusiform gyrus irrespective of task
complicated figures are presented and condition. HC showed greater precuneus
the participant is required to indicate activation than AN.
which of the two figures contains the
target shape.

(Continued)

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 5 31/10/2013 10:22:17 AM

 6 ANZJP Articles

Table 3. (Continued)

Authors Participants Procedure Key findings

Lao-Kaim et al. 31 AN (24 AN-R, Participants completed a letter n-back AN vs HC: no difference in accuracy or
(2013) 7 AN-BP), 31 HC task. response time at any task difficulty; no
significant difference in activations at each
level of difficulty.

Oberndorfer et al. 14 rec-AN, 14 Sucrose and sucralose solutions Rec-AN vs HC: sucrose and sucralose
(2013) rec-BN, 14 HC were delivered to participants while resulted in decreased right anterior insula
undergoing functional magnetic activation.
resonance imaging.

Sato (2013) 15 AN (9 AN-R, 6 The Wisconsin Card Sort Test was AN vs HC: lower right ventrolateral
AN-BP), 15 HC administered during scanning. prefrontal cortex and bilateral
parahippocampal cortex activity during set
shifting.

Strigo et al. (2013) 12 rec-AN, 12 HC Painful heat stimuli were administered Rec-AN vs HC: greater activity of the
and different colours indicated the right anterior insula, DLPFC and cingulate
intensity of the upcoming trial. cortex during pain anticipation; increased
activity during pain stimulation in the
DLPFC and decreased activity in the
posterior insula.

Suchan et al. 10 AN, 15 HC Passive viewing of images of bodies AN vs HC: effective connectivities
(2013) and chairs. between the middle occipital gyrus and
the fusiform body area and the fusiform
body area and the extrastriate body area;
reduced connectivity of the fusiform body
area and extrastriate body area.

Suda (2013) 20 AN, 15 HC Participants were presented with AN vs HC: reduced activity of the
images of ‘body checking’ and control anteromedial prefrontal cortex and right
images of other active tasks, and fusiform gryus. The prefrontal cortex
were asked to imagine they were activity was negatively correlated with
performing the action portrayed in shape concern; increased activation of the
the image. right parietal lobe to the body checking
images compared to the neutral images in
AN.

Bar et al. (2012) 19 AN-R, 19 HC Pain processing was investigated by AN vs HC: increased activity in the
delivering thermal painful stimuli to ipsilateral pons during heat pain
the right arm. perception.

Brooks et al. 18 AN (11 AN-R, Presented with high- and low-calorie AN vs HC: increased activity of the right
(2012) 7 AN-BP), 24 HC food and non-food images and visual cortex, and reduced activity of the
required to imagine eating the food or bilateral cerebellar vermis when thinking
using the non-food items. about eating food.

Castellini et al. 18 AN-R, 19 HC Passive viewing of images of their own AN vs HC: increased activity in the middle
(2012) body and digitally distorted images of temporal gyrus in the underweight–normal
their own body as underweight and weight contrast, and increased inferior
overweight. Control stimuli were frontal gyrus activity in the overweight–
images of houses. normal weight comparison.

Cowdrey et al. 16 rec-AN, 15 HC Resting state scans were performed Rec-AN vs HC: increased temporal
(2012a) while participants had their eyes open. coherence between the default mode
Resting state networks of interest network functional connectivity map and
included medial visual, lateral visual, the right precuneus close to the border
auditory, sensory-motor, default of the posterior cingulate gyrus and the
mode network, cognitive control, and prefrontal cortex/inferior frontal gyrus.
fronto-parietal (left and right).
(Continued)

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 6 31/10/2013 10:22:17 AM

 Phillipou et al. 7

Table 3. (Continued)

Authors Participants Procedure Key findings

Cowdrey et al. 16 rec-AN-R, 16 Presented with happy and fearful No difference between groups.
(2012b) HC emotional faces and required to
undertake a gender-decision task.

Favaro et al. 33 AN, 30 HC Resting state scans were performed AN vs HC: groups did not differ in functional
(2012a) while participants had their eyes connectivity of the prefrontal cortex.
closed. AN MET homozygotes vs AN Val carriers:
higher positive coactivation of the DLPFC
and ventromedial prefrontal cortex seeds,
and lower negative coactivation for the
DLPFC, ventromedial prefrontal cortex
and ventrolateral prefrontal cortex.

Favaro et al. 29 AN, 16 rec- Resting state scans were performed AN vs HC: decreased connectivity
(2012b) AN, 26 HC while participants had their eyes in the left occipitotemporal junction;
closed. Three visual networks significantly increased coactivation within
(medial, lateral and ventral) and the somatosensory network in the left
one somatosensory network were superior parietal cortex.
identified. Rec-AN vs HC: decreased activation in the
right middle frontal gyrus.

Frank et al. (2012) 21 AN-R, 19 Participants performed a reward- AN-R vs HC: greater right ventral
obese, 23 HC conditioning task that involved putamen activation.
learning the association between AN-R vs HC and obese: greater left
conditioned visual stimuli (geometric orbitofrontal cortex activation in the
shape) and unconditioned taste positive prediction error condition; reduced
stimuli (sucrose solution, no solution activation of the bilateral putamen and
or artificial saliva), and unexpected left orbitofrontal cortex for the negative
violations of the learned associations. prediction error condition.

Holsen et al. 12 AN-R, 10 rec- Presented with images of high- and AN vs HC: reduced hypothalamus,
(2012) AN (8 AN-R, 2 low-calorie foods and objects amygdala, hippocampus, anterior insula and
AN-BP), 11 HC (control) before and after a meal. orbitofrontal cortex activity to high-calorie
Participants were asked to press a foods pre-meal; reduced amygdala and
button when images changed. insula activity post-meal.
Rec-AN vs HC: decreased activity of the
hypothalamus, amygdala and anterior insula
pre-meal.

Kim et al. (2012) 18 AN (6 AN-R, Passive viewing of food and non-food AN vs HC: increased activity of the left
12 AN-BP), 20 images. anterior insula and significant interactions
BN, 20 HC with the right insula and inferior frontal gyrus.

McAdams et al. 18 rec-AN, 18 HC Presented with written appraisal Rec-AN vs HC: self-knowledge and
(2012) statements in social and physical tasks perspective-taking showed decreased
that required the evaluation of the activity of the precuneus, and reduced left
self, a friend and a reflection of what is middle frontal gyrus activation.
believed by one’s friend.

Miyake et al. 30 AN, 20 HC Presented with negative words AN vs HC: higher activations in the left
(2012) concerning interpersonal relationships insula, and right superior temporal and
and neutral words. inferior frontal gyri.

Brooks et al. 18 AN (11 AN-R, Presented with high- and low- AN-R vs HC: increased activation of the
(2011) 7 AN-BP), 8 BN, calorie food and non-food images. cerebellum, left visual cortex, right DLPFC
24 HC Participants were required to imagine and parietal lobe to food images compared
eating the food or using the non-food to non-food images.
items. AN-BP vs HC: increased activity in the
bilateral cerebellum and supplementary
motor area in to food images.
AN vs HC: results not presented.
(Continued)

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 7 31/10/2013 10:22:17 AM

 8 ANZJP Articles

Table 3. (Continued)

Authors Participants Procedure Key findings

Cowdrey et al. 15 rec-AN-R, 16 Participants were delivered an oral Rec-AN-R vs HC: increased activation
(2011) HC stimulus of either a pleasant chocolate of the insula, cingulate, occipital and
taste or an aversive unpleasant medial prefrontal cortices to the taste
taste during the visual presentation of chocolate; increased activation of the
of a picture of chocolate, mouldy ACC, lateral posterior insula, caudate,
strawberries or a grey control image, putamen and DLPFC in response to the
or a tasteless liquid coupled with a aversive image and taste condition.
grey screen (control).

Joos et al. (2011) 11 AN-R, 11 HC Passive viewing of food and non-food AN-R vs HC: decreased activity of the
items. posterior middle cingulate cortex and
increased activity of the right amygdala to
food images.

Oberndorfer et al. 12 rec-AN, 12 HC Completed a stop signal task. AN vs HC: similar activity between groups
(2011) when inhibitory demand is low; reduced
medial prefrontal cortex activity during
hard trials.

Vocks et al. 12 AN-R, 12 HC Participants drank chocolate milk and AN-R vs HC: increased activations when
(2011) water via a tube in hungry and satiated drinking chocolate milk in the right
states. amygdala and left medial temporal gyrus
when hungry; no differences in the satiated
condition; increased activity of the inferior
temporal gyrus (including the extrastriate
body area), whereas HC showed increased
activation of the left insula when comparing
the hungry and satiated conditions.

Castro-Fornieles T1: 14 AN, 14 HC Testing was completed during T1: increased activation of temporal and
et al. (2010) T2: 9 rec-AN admittance and after 7 months of parietal areas in AN.
treatment and weight recovery. T2: rec-AN showed reduced activity of
Participants complete a number the ACC, and frontal and parietal regions
n-back task. compared to T1, but did not differ in
activations relative to HC.

Fladung et al. 14 AN, 14 HC Presented with computer-generated AN vs HC: increased activity to

(2010) images of underweight, overweight underweight relative to normal weight
and normal weight female bodies. stimuli; increased ventral striatal activity
Asked to process stimulus in a self- to normal weight stimuli compared to
referring way or estimate body weight underweight stimuli in HC.
(control).

Friederich et al. 17 AN, 18 HC Compare self with slim idealised AN vs HC: increased activity of the right
(2010) female bodies (active condition) or insula and premotor cortex, and less
compare their room design with activity of the rostral ACC in response to
images of interior home designs the active condition.
(control condition).

Gizewski et al. 12 AN-R, 12 HC Presented with images of high-calorie AN-R vs HC: increased activity of the
(2010) foods and neutral stimuli in satiated central cortices and to food stimuli in
and hungry states. the hungry state, and the ACC and insula
in the HC; increased posterior cingulate
cortex activation in the hungry state;
increased activity of the left insula during
the satiated state.

Miyake et al. 12 AN-R, 12 AN- Presented with negative body-image AN-R and AN-BP vs BN and HC: higher
(2010) BP, 12 BN, 12 HC words, non-specific emotional words activation of the right amygdala to negative
and neutral words. body-image words.
(Continued)

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 8 31/10/2013 10:22:18 AM

 Phillipou et al. 9

Table 3. (Continued)

Authors Participants Procedure Key findings

AN-R and AN-BP vs HC: increased left IPL

activity.
AN-BP and BN vs HC: increased left
ventromedial prefrontal cortex.

Mohr et al. (2010) 16 AN, 16 HC Images of the individual’s body were AN vs HC: increased activity of the insula
created that were thinner and fatter and middle frontal gyrus for thin images for
than their actual body. Task 1: rate Task 2 compared to thin images of Task 1;
how much image corresponds to their lower activation of the precuneus for all
real body. Task 2: rate how much body sizes in Task 1, and the fatter images
image represents their ideal body. for Task 1 relative to Task 2.

Vocks et al. 13 AN, 15 BN, Passive viewing of images of their own AN vs HC: reduced activity of the inferior
(2010a) 27 HC body and another woman’s body. parietal lobule when viewing their own body.
AN vs BN and HC: higher amygdala
activity.

Vocks et al. 5 AN Passive viewing of their own body. Increased activity pre- to post-treatment in
(2010b) Tested before and after body image the right middle temporal gyrus (including
therapy. the extrastriate body area). Decreased
activity pre- to post-treatment in the
left precuneus, posterior cingulate gyrus,
fusiform gyrus, right inferior and superior
frontal gyri, parahippocampal gyrus and
bilateral inferior parietal lobule.

Zastrow et al. 15 AN (7 AN-R, 8 Presented with a visual target- AN vs HC: reduced activity of the dorsal
(2009) AN-BP), 15 HC detection task consisting of images of ACC and ventral putamen in the correct
geometric shapes. Participants were target shift trials; reduced activity of
required to classify shapes as target, the rostral ACC and middle and dorsal
non-target or standard shapes. striatum in the target maintain trials;
incorrect trials, relative to correct trials,
resulted in decreased activity of the dorsal
ACC extending to the medial frontal gyrus.
AN: decreased activity in the bilateral
thalamus, rostral cingulate, sensorimotor
region and cerebellum to target trials
compared to standard trials, independent
of task performance.

Redgrave et al. 6 AN, 6 HC Emotional Stroop task consisting of AN vs HC: greater activity of the middle
(2008) fat, thin and neutral words and words and medial frontal gyri, and the junction
consisting of XXXXs. of the insula, frontal and temporal lobes in
the thin-XXXX contrast; reduced activity
of the DLPFC in the fat-XXXX contrast.

Sachdev et al. 10 AN, 10 HC Presented with self and non-self body AN vs HC: less activity in the frontal gyri,
(2008) images in a passive viewing task. insula, precuneus and occipital regions
when processing self-images but did not
differ when processing non-self images.

Wagner et al. 16 rec-AN-R, 16 Presented with a sucrose solution or Rec-AN-R vs HC: reduced activation to
(2007a) HC distilled water. both tastes in the insula, and ventral and
dorsal striatum.

Wagner et al. 13 rec-AN-R, 13 Completed a simple monetary reward Region-of-interest analysis: rec-AN-R showed
(2007b) HC task. decreased activity in the caudate than HC.
HC showed increased activity in the left
anterior ventral striatum for wins than for
losses, whereas rec-AN-R did not differ.
(Continued)

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 9 31/10/2013 10:22:18 AM

 10 ANZJP Articles

Table 3. (Continued)

Authors Participants Procedure Key findings

Voxel-wise image analysis: rec-AN-R

showed increased activation in the anterior
and dorsal striatum to a win compared to
a loss.

Santel et al. (2006) 13 AN, 10 HC Participants rated food and non-food AN vs HC: decreased activity of the left
images for pleasantness in hungry and inferior parietal cortex when satiated and
satiated states. the right visual occipital cortex to food
images when hungry; compared to the
hungry state, satiety in AN was associated
with increased activity of the right occipital
cortex.

Uher et al. (2005) 13 AN, 9 BN, 18 Presented with line drawings of AN vs BN and HC: weaker activation of
HC underweight, overweight and normal the occipitotemporal cortex (including the
weight female bodies, and houses extrastriate body area), fusiform gyrus and
(control), and asked to think how parietal cortex to all body shapes.
acceptable the body/house would be
for them.

Uher et al. (2004) 16 AN, 10 BN, Presented with images of food, and AN vs HC: more activity in the lingual
19 HC non-food items and asked to think gyrus and reduced activity in the inferior
how hungry it makes them; and parietal lobule and cerebellum, in the
emotional aversive and neutral images food–non-food contrast; reduced activity
and asked to think about how it in the right cerebellum but increased
makes them feel. activity of the left cerebellum in the
aversive-neutral contrast.

Uher et al. (2003) 8 AN-R, 9 rec- Presented with images of food and Rec-AN-R vs HC: increased medial
AN-R, 9 HC non-food items and asked to think prefrontal and ACC, and reduced inferior
how hungry it makes them; and parietal lobule activity in response to food
emotional aversive and neutral images versus non-food images.
and asked to think about how it Rec-AN-R vs AN-R: increased activation
makes them feel. of dorsal ACC, right lateral and apical
prefrontal cortex.
No difference in emotional aversive and
neutral comparison between groups.

Wagner et al. 13 AN, 10 HC Presented with distorted images AN vs HC: increased activity of the
(2003) of their own body with subjective inferior parietal lobule and prefrontal
maximum unacceptability (target), cortex in relation to target versus neutral
subjective unacceptability of another stimuli.
woman’s body (non-target), and
abstract images consisting of
participant’s own body (neutral).

Seeger et al. 3 AN, 3 HC Presented with distorted images AN vs HC: target versus non-target and
(2002) of their own body with subjective neutral contrast revealed increased activity
maximum unacceptability (target), of the right amygdala, fusiform gyrus and
subjective unacceptability of another brainstem.
woman’s body (non-target), and
abstract images consisting of
participant’s own body (neutral).

Ellison et al. 6 AN, 6 HC Presented with images of high- and AN vs HC: increased activity of the left
(1998) low-calorie drinks. insula, ACC and left amygdala-hippocampal
region to high- versus low-calorie drinks.

AN: anorexia nervosa; AN-R: anorexia nervosa restricting subtype; AN-BP: anorexia nervosa binge-purge subtype; rec-AN: weight-recovered AN;
BN: bulimia nervosa; HC: healthy controls; DLPFC: dorsolateral prefrontal cortex; ACC: anterior cingulate cortex; T1: time 1; T2: time 2; T3: time 3.

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 10 31/10/2013 10:22:18 AM

 Phillipou et al. 11

Table 4.  Functional brain differences in anorexia nervosa utilising modalities other than functional magnetic resonance imaging.

Modality Authors Participants Procedure Key findings

PET

Bailer et al. 10 rec-AN, 9 Participants were scanned pre- Rec-AN vs HC: positive association
(2012) HC amphetamine administration between anxiety and DA release in
and 3 hours post-amphetamine the dorsal caudate.
administration.

  Galusca 8 AN-R, 9 rec- All participants were tested once, AN-R vs HC: increased 5-HT1A
et al. (2008) AN-R, 7 HC 2 hours after lunch. Recovery was receptor binding in the right superior
defined as a BMI > 18.5 kg/m2 for at temporal gyrus, inferior frontal
least 1 year. gyrus, parietal operculum and
temporaparietal junction.

  Bailer et al. 15 AN, 29 HC All participants were tested once. AN vs HC: increased 5-HT1A
(2007a) receptor binding in prefrontal and
lateral orbitofrontal regions, mesial
and lateral temporal lobes, parietal
cortex, and dorsal raphae nuclei.
5-HT2A was normal.

  Bailer et al. 8 rec-AN-R, 7 All participants were tested once. Rec-AN-R vs rec-AN-BP: increased
(2007b) rec-AN-BP, 9 Recovery was defined as having 5-HT binding potential in the dorsal
rec-BN, 10 HC maintained a weight above 85% raphe and anteroventral striatum.
average body weight, regular Rec-AN-R vs rec-BN: decreased
menstrual cycles, and not having 5-HT binding potential in the
binged, purged, or engaged in anteroventral striatum.
significant restrictive eating patterns ED vs HC: no differences.
for at least 1 year before the study.

  Bailer et al. 13 rec-AN-R, All participants were tested once. Rec-AN-BP vs HC: increased 5-HT1A
(2005) 12 rec-AN-BP, binding in cingulate, lateral and mesial
18 HC temporal, lateral and medial orbital
frontal, parietal, and prefrontal
cortical regions, and dorsal raphe.
AN-R vs HC: no significant
differences.

  Frank et al. 10 rec-AN, 12 All participants were tested during Rec-AN vs HC: higher DA D2/D3
(2005) HC the first 10 days of the follicular phase receptor binding in antero-ventral
of the menstrual cycle. Recovery was striatum.
defined as 1 year of normal weight,
regular menstrual cycles and no
bingeing or purging behaviours.

  Bailer et al. 10 rec-AN-BP, All participants were tested during Rec-AN-BP vs HC: had significantly
(2004) 16 HC the first 10 days of the follicular phase reduced 5-HT2A binding potential the
of the menstrual cycle. Recovery was left subgenual cingulate, left parietal
defined as 1 year of normal weight, cortex, and right occipital cortex.
regular menstrual cycles and no
bingeing or purging behaviours.

  Frank et al. 16 rec-AN-R, All participants were tested once. Rec-AN-R vs HC:had significantly
(2002) 23 HC Recovery was defined as 1 year of reduced 5-HT2A binding in the
normal weight, regular menstrual cycles amygdala, hippocampus and cingulate.
and no bingeing or purging behaviours.

  Frank et al. 10 rec-AN-R, All participants were tested once. rCBF was similar across groups.
(2007) 8 rec-AN-BP, 9 Recovery was defined as 1 year of
rec-BN, 8 HC normal weight, regular menstrual cycles
and no bingeing or purging behaviours.

(Continued)

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 11 31/10/2013 10:22:18 AM

 12 ANZJP Articles

Table 4. (Continued)

Modality Authors Participants Procedure Key findings

  Gordon 8 AN, 8 HC Participants were presented with high- AN vs HC: greater rCBF in the
et al. (2001) and low-caloric foods, and non-food medial temporal lobe; increased
items. activity of the left occipital and right
temporo-occipital cortex for high-
calorie images relative to low-calorie
images.

  Delvenne 10 AN, 10 BN, All participants were tested once. AN vs BN and HC: lower absolute
et al. (1999) 10 HC global cortical glucose activity, and
higher relative activity in the inferior
frontal cortex, basal ganglia, putamen
and caudate.

  Delvenne 10 AN, 10 All participants were tested once. AN vs HC: absolute global and
et al. (1997) low-weight regional glucose activity was lower;
depressive lower relative metabolism of glucose
disorder, 10 in the parietal cortex.
HC

  Delvenne T1: 10 AN, 10 All participants were scanned during T1: AN participants demonstrated a
et al. (1996) HC rest, eyes closed and with low ambient regional hypometabolism of glucose.
T2: 10 rec-AN noise. The AN participants were T2: Regional hypometabolism of
scanned on two occasions, when ill glucose normalised with weight gain.
(and underweight) and when weight-
recovered

  Delvenne 10 AN, 10 HC All participants were scanned during AN vs HC: regional hypometabolism
et al. (1995) rest, eyes closed and with low ambient of glucose in cortical regions, with the
noise. most significant differences found in
the frontal and the parietal cortices.

  Herholz T1: 5 AN, 15 AN participants were tested during T1: AN had significant
et al. (1987) HC the ill AN state and following weight hypermetabolism of the caudate,
T2: 5 rec-AN recovery. temporal cortex, lentiform nucleus,
thalamus and brainstem.
T2: When weight-recovered, there
were no differences in comparison
to HC.

SPECT  

  Frampton T1: 13 AN Participants were tested when ill T1: unilateral temporal
et al. (2011) T2: 9 rec-AN and when weight-recovered, with an hypoperfusion, predominantly in the
average of 4.2 years following the first antero-medial temporal region.
session. T2: hypoperfusion persisted.

  Komatsu T1: 10 AN Participants were tested when ill, and Increased rCBF in the bilateral
et al. (2010) T2: 10 AN after weight loss and normal eating parietal lobe and right posterior
behaviours improved with an average cingulate gyrus after weight gain.
duration of 120 days following the first
session.

  Beato- 9 AN-R, 13 Results were compared during a AN vs HC: increased activity of

Fernandez BN, 12 HC rest condition, following a video of a the left parietal and right superior
et al. (2009) landscape (neutral), and following a frontal cortices when viewing one’s
video of the participant’s own body own body compared to the neutral
filmed unclothed. condition.

  Yonezawa 13 AN-R, 13 All participants were tested once. AN-R and AN-BP vs HC: decreased
et al. (2008) AN-BP, 10 HC perfusion bilaterally of subcallosal gyrus,
midbrain and posterior cingulate.

(Continued)

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 12 31/10/2013 10:22:18 AM

 Phillipou et al. 13

Table 4. (Continued)

Modality Authors Participants Procedure Key findings

  Key et al. 11 AN, 11 HC All participants were tested once. Hypoperfusion was demonstrated
(2006) in the anterior temporal lobe
and/or caudate nuclei in 8 of
11 AN patients. There was no
hypoperfusion in HC.

  Matsumoto 8 AN All participants were tested within 2 rCBF increases post-treatment were
et al. (2006) weeks of hospital admission and prior observed in the DLPFC, medial
to being discharged. prefrontal cortex, posterior and
ACC and the precuneus.

  Kojima et al. T1: 12 AN-R, Control participants were tested T1: AN-R had lower rCBF in the
(2005) 11 HC once, whereas AN participants were bilateral anterior lobes, including the
T2: 12 rec- retested following weight recovery. ACC, right parietal lobe, insula and
AN-R occipital lobes.
T2: AN-R showed increases in the
right parietal lobe, and decreases
in the basal ganglia, cerebellum and
ACC compared to HC.

  Audernert 15 AN, 11 HC All participants were tested once. Significantly reduced 5-HT2A binding
et al. (2003) potential in the left frontal cortex,
left and right parietal cortex, and left
and right occipital cortex.

  Chowdhury 15 AN All participants were tested once. 11 of 15 participants showed

et al. (2003) hypoperfusion in at least one area
including the temporal lobe, parietal
lobe, frontal lobe, thalamus and the
caudate nuclei.

  Naruo et al. 7 AN-R, 7 AN- All participants were tested once. AN-R vs AN-BP and HC:
(2001) BP, 7 HC hypoperfusion of frontal areas,
mainly the ACC.

  Takano 14 AN, 8 HC All participants were tested once. AN vs HC: had hypoperfusion in
et al. (2001) the medial prefrontal cortex and
ACC, and hyperperfusion in the
thalamus and amygdala hippocampus
complex.

  Naruo et al. 7 AN-R, 7 AN- All participants were tested once AN-BP vs HC: increased rCBF in
(2000) BP, 7 HC before breakfast. the right inferior prefrontal,
superior prefrontal and parietal
regions.

  Gordon T1: 15 AN 15 participants were tested when T1: 13 of the 15 patients

et al. (1997) T2: 3 rec-AN ill with childhood AN; 3 of these had unilateral temporal lobe
participants were retested after weight hypoperfusion.
restoration. T2: temporal lobe hypoperfusion
persisted in the three participants
tested at follow-up.

  Nozoe et al. 8 AN, 5 BN, SPECT was performed before and AN vs BN and HC: rCBF activity
(1994) 9 HC after food intake. was lower in all cortical regions
pre-meal.
AN vs HC: left parietal rCBF was
significantly lower pre-meal.
Post-meal, the three groups did not
differ.

(Continued)

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 13 31/10/2013 10:22:18 AM

 14 ANZJP Articles

Table 4. (Continued)

Modality Authors Participants Procedure Key findings

  Kuruoglu T1: 2 AN 2 individuals with AN were tested T1: bilateral hypoperfusion in frontal,
et al. (1985) T2: 2 AN when ill and following treatment. parietal and frontotemporal areas.
following T2: normal brain perfusion.
treatment

Lumbar puncture  

  Kaye et al. 19 rec-AN, 13 All participants were tested during the AN-R vs AN-BP, BN and HC: had
(1999) rec-BN, 19 HC first 10 days of the follicular phase of significantly lower CSF HVA.
the subject’s menstrual cycle

  Kaye et al. 17 rec-AN, 15 All participants were tested during the Rec-AN vs HC: elevated
(1991) HC early follicular phase of the menstrual concentrations of CSF increased
cycle. 5-HIAA levels; no difference was
found for CSF HVA levels.

  Kaye et al. T1: 15 AN, 10 All participants were tested during the T1: CSF 5-IAA levels were
(1988) HC early follicular phase of the menstrual significantly reduced in AN compared
T2: 11 AN cycle, between 8am and 9am. to HC.
following T2: CSF 5-IAA levels increased in
2–6 weeks AN.
of a weight-
restoration
programme

  Kaye et al. 8 AN (5 AN- All participants were administered a AN-BP vs AN-R: no difference in
(1984a) BP, 3 AN-R), lumbar puncture between 8am and CSF 5-HIAA or HVA.
8 recently 9am (pre-probenecid). Probenecid was Long-term rec-AN vs recent rec-
rec-AN (5 AN- infused over 15 minutes and another AN: higher post-probenecid minus
BP, 3 AN-R), lumbar puncture was performed 8 pre-probenecid CSF HVA; similar
8 long-term hours later (post-probenecid). 5-HIAA.
rec-AN (6 AN- Underweight AN participants were Combined rec-AN vs HC: no
BP, 2 AN-R), continuously below 75% of ideal body difference in CSF HVA pre- or post-
8 HC weight for a minimum of 6 months; the probenecid; no difference in CSF
same patients were tested 3–4 weeks 5-HIAA pre-probenecid; rec-AN-BP
after weight correction (recently rec- had higher 5-HIAA than AN-R or
AN); and a separate group with AN HC post-probenecid.
who were weight-recovered with a
mean of 20 months (long-term rec-AN).

  Kaye et al. T1: 16 AN (5 All participants were administered a T1: ill AN had lower CSF 5-HIAA
(1984b) AN-R, 11 AN- lumbar puncture between 8am and and HVA.
BP); 8 long- 9am, and a second following 8 hours. T2: CSF, 5-HIAA and HVA
term rec-AN returned to normal with weight
T2: 8 of the ill restoration. Rec-AN-R had a higher
AN following concentration of 5-HIAA than rec-
3–4 weeks AN-BP.
of weight
recovery

Blood samples  

  Brambilla 8 AN-R, 8 All participants were tested once. Reduced D2 receptor sensitivity in
et al. (2001) AN-BP,7 BN, AN-R compared to other groups.
8 HC

AN: anorexia nervosa; AN-R: anorexia nervosa restricting subtype; AN-BP: anorexia nervosa binge-purge subtype; rec-AN: weight-recovered AN;
BN: bulimia nervosa; ED: eating disorders; HC: healthy controls; MRI: magnetic resonance imaging; SPECT:single photon emission computed tomog-
raphy; Tc: technetium; HMPAO: hexamethylpropylene amine oxime; rCBF: regional cerebral blood flow; PET: positron emission tomography; 5-HT:
serotonin; DA: dopamine; HVA: homovanillic acid; 5-HIAA: 5-hydroxyindoleacetic acid; CSF: cerebrospinal fluid; BMI: body mass index; T1: time 1;
T2: time 2; T3: time 3; DLPFC: dorsolateral prefrontal cortex; ACC: anterior cingulate cortrex.

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 14 31/10/2013 10:22:18 AM

 Phillipou et al. 15

patients and patients with a current diagnosis. Furthermore, Decreases were also reported in the left superior longitudi-
in comparison to weight-recovered individuals, individuals nal fasciculus, bilateral posterior coronal radiate and left
ill with AN were found to have reduced grey matter vol- middle cerebellar peduncle. Additionally, no differences
umes of the left amygdala and putamen, and the bilateral were found between the two subgroups of AN, currently ill
inferior temporal cortex. Frank et al. (2013) on the other or weight-recovered.
hand reported increased grey matter volume of the gyrus
rectus and decreased insula volume compared to healthy
controls. This study also reported reduced grey matter vol- Functional differences
ume of the caudate and putamen and reduced inferior pari-
Neuronal systems
etal white matter volume in weight-recovered AN, as well
as reduced temporal white matter volume in both weight- A neuronal system of particular interest in relation to the
recovered and ill AN. Other recent studies have also found neurobiology of AN is the serotonin (5-HT) system.
decreased grey matter volumes in ill AN patients of the Research in AN has mainly focused on the main metabolite
extrastriate body area (Suchan et al., 2010), and the middle of 5-HT, 5-hydroxyindoleacetic acid (5-HIAA), and the
cingulate cortex, precuneus, and inferior and superior pari- 5-HT receptors 5-HT1A and 5-HT2A. Individuals with AN in
etal cortex (Gaudio et al., 2011). A recent study by Amianto the ill state have significantly lower CSF basal concentra-
et al. (2013) reported decreased grey matter in the lateral tions of the 5-HT metabolite 5-HIAA (Kaye et al., 1984b,
cerebellum, precuneus and frontal orbital and cingulate 1988). However, those recovered from AN have signifi-
cortices. Furthermore, a study by Boghi et al. (2011) found cantly increased 5-HIAA levels in comparison to healthy
reduced grey matter of the frontal and parietal cortices, cau- controls (Kaye et al., 1991), particularly those recovered
date nucleus, hypothalamus and cerebellum. Additionally, from the binge-eating purging subtype of AN (AN-BP)
in this study the AN group was divided into two subgroups (Kaye et al., 1984a).
based on illness duration: those who had a diagnosis of Recent neuroimaging studies have investigated 5-HT
more than nine years, and those who were being treated at binding differences in the brain. In a study undertaken by
first presentation of the illness. The group with shorter dis- Bailer and colleagues (2007b), the investigators reported
ease duration were found to have more evident hypotha- that ill AN individuals had increased 5-HT1A binding in the
lamic atrophy, whereas the cerebellum was more affected subgenual, mesial temporal, orbital frontal and raphae brain
in the group with longer disease duration, in comparison to regions, and the prefrontal, lateral temporal, ACC and pari-
healthy controls. Additionally, individuals with AN have etal regions. Similar findings have been reported for 5-HT1A
been found to have reduced hippocampal volumes in com- binding in ill participants (Bailer et al., 2007b; Galusca
parison to healthy controls (Connan et al., 2006), and the et al., 2008), whereas recovered AN sufferers have been
size of the mesencephalon has also been found to be mark- reported to have diminished binding potential for 5-HT2A
edly reduced in ill AN patients, which remains reduced in and increased binding potential of 5-HT1A. Audenaert et al.
weight-restored individuals (Neumärker et al., 2000). (2003) reported significantly reduced 5-HT2A binding in
Several recent studies have also utilised diffusion tensor the left frontal cortex as well as the left and right parietal
imaging (DTI) to assess white matter pathways in AN. and occipital cortices of individuals with AN in comparison
Kazlouski et al. (2011) reported poorer white matter integ- to control participants. Bailer et al. (2005), on the other
rity in the bilateral fimbira-fornix, fronto-occipital fascicu- hand, reported increased 5-HT1A binding potential of the
lus and posterior cingulum in AN, irrespective of AN dorsal raphe, and the cingulate, lateral and medial tempo-
subgroup. The AN group also showed higher mean diffu- ral, lateral and medial orbitofrontal, parietal and prefrontal
sivity in these areas, in addition to frontal, parietal, tempo- cortices in weight-recovered AN. Studies investigating the
ral and occipital areas. However, in a more recent study by specific subtypes of AN found that weight-restored restrict-
Yau et al. (2013), the authors reported lower mean diffusiv- ing subtype AN (AN-R) participants had reduced 5-HT2A
ity in frontal, parietal and cingulum white matter tracts in receptor activity in the subgenual and pregenual cingulate
weight-recovered AN, unlike the study by Kazlouski et al. cortex, and mesial temporal and parietal cortical areas
(2011) who found higher mean diffusivity in these areas. (Frank et al., 2002), whereas reduced 5-HT2A receptor
Yau et al. (2013) reported poorer white matter integrity in activity in the subgenual cingulate, and the mesial tempo-
weight-recovered AN who had a history of more severe ill- ral, lateral temporal, parietal and occipital cortices has been
ness. Frieling et al. (2012), on the other hand, investigated reported for recovered AN-BP participants, in comparison
white matter integrity in a group of individuals with a cur- to controls (Bailer et al., 2004). A further study, however,
rent diagnosis of AN combined with individuals weight- found no difference in 5-HT binding potential between AN
recovered from the illness, and a sample of healthy subgroups and controls, but increased 5-HT binding poten-
participants. This study reported regional decreases in tial in the dorsal raphe and anteroventral striatum in recov-
white matter of the posterior thalamic radiation bilaterally ered AN-R compared to recovered AN-BP participants; and
and the mediodorsal thalamus in the combined AN group. decreased 5-HT binding potential in the anteroventral

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 15 31/10/2013 10:22:18 AM

 16 ANZJP Articles

striatum in AN-R compared to recovered individuals with neuropeptide-Y and leptin, although the differences observed
BN (Bailer et al., 2007a). appear to normalise after weight restoration, suggesting a
A neurotransmitter system largely related to the seroton- state rather than trait effect (for a review see Bailer and Kaye,
ergic system which has also been implicated in the aetiol- 2003).
ogy and course of AN is the dopaminergic system.
Homovanillic acid (HVA), the major metabolite of dopa-
Regional cerebral blood flow
mine (DA) in humans, has been found to be decreased in
the CSF of individuals with AN in the ill state (Kaye et al., Functional neuroimaging is relatively recent technology
1984b). Two related studies have reported that HVA returns which has allowed the indirect measurement of brain activ-
to normal levels following weight restoration (Kaye et al., ity. Early functional studies utilised single-photon emission
1984a, 1984b). However, a more recent study undertaken computed tomography (SPECT) and showed hypo- and
by the same investigators examining the HVA levels in hyperperfusion at rest in a number of brain areas in the ill
individuals recovered from AN utilising more stringent cri- state of AN. Kuruoglu et al. (1985) reported hypoperfusion
teria than purely weight restoration (including over one in frontal, parietal and frontotemporal areas, which normal-
year of normal weight, regular menstrual cycles, no restrict- ised following weight restoration. Another study reported
ing eating patterns and no bingeing or purging) found that unilateral temporal lobe hypoperfusion which persisted in a
individuals recovered from AN-R had significantly lower subgroup who had a follow-up scan following weight resto-
CSF HVA than AN-BP, bulimia nervosa (BN) participants ration (Gordon et al., 1997). A more recent study by
and healthy controls (Kaye et al., 1999). Chowdhury et al. (2003) revealed hypoperfusion in the
In related work, Frank et al. (2005) used positron emis- temporal, parietal and frontal lobes, thalamus, and caudate
sion tomography (PET) imaging with radiogland [11C] nuclei. Other recent studies found hypoperfusion in the
raclopride to assess DA D2/D3 receptor function in recov- medial prefrontal cortex, the ACC (Takano et al., 2001), the
ered AN individuals. The investigators reported signifi- posterior cingulate gyrus, the subcallosal gyrus, the mid-
cantly higher [11C]raclopride binding potential in the brain (Yonezawa et al., 2008), the anterior temporal lobe
antero-ventral striatum than control participants, providing and caudate nuclei (Key et al., 2006), and hyperperfusion in
support for the possibility that AN is associated with the thalamus and the amygdala-hippocampus complex
decreased intrasynaptic DA concentration or increased D2/ (Takano et al., 2001). Hypoperfusion specific to AN-R has
D3 receptor density; the authors suggest this may contrib- also been found in frontal areas, mainly the ACC (Naruo
ute to the disturbed reward processing exhibited in AN. et al., 2001), and hyperperfusion specific to AN-BP in the
Another study using an amphetamine challenge and PET right hemisphere, including inferior and superior prefron-
found that recovered AN participants exhibited a positive tal, and parietal regions (Naruo et al., 2000). Changes in
association between endogenous DA release and anxiety in regional cerebral blood flow (rCBF) at rest have been
the dorsal caudate, possibly explaining why food-related reported by some to normalise following weight restoration
DA release produces anxiety in AN but is pleasurable in and remission (Kuruoglu et al., 1985), though others have
healthy individuals (Bailer et al., 2012). found persistent changes (Frampton et al., 2011). One study
Additionally, AN has been associated with altered D2 reported that, in comparison with healthy controls, indi-
receptor sensitivity demonstrated by significantly reduced viduals with AN-R prior to treatment had lower cerebral
growth hormone response to growth hormone releasing blood flow in the ACC, right parietal lobe, insula and occip-
hormone administered with apomorphine, a selective D1 ital lobes. Following treatment and weight gain, normalisa-
and D2 receptor agonist (Brambilla et al., 2001). A study tion in a number of brain areas occurred, but decreased
indirectly investigating dopaminergic function in AN also rCBF persisted in the ACC (Kojima et al., 2005). Another
revealed a deficit in this group (Lawrence et al., 2003). study by Matsumoto et al. (2006) found AN patients had
Participants were administered a task which involved the increased rCBF post-treatment relative to pre-treatment in
learning of a series of two-alternative force-choice visual the dorsolateral prefrontal cortex (DLPFC), medial pre-
discriminations. This is analogous to tasks that activate DA frontal cortex, anterior and posterior cingulate and precu-
neurons in primates and is sensitive to neurotransmission, neus. A more recent study by Komatsu et al. (2010),
including L-dopa treatment in Parkinson’s disease. The AN however, reported increased rCBF in the right posterior
group was found to show deficits in learning during the cingulate gyrus and bilateral parietal lobe following weight
early stages of the task when DA activity should be at a recovery. Differences in rCBF have also been reported in
maximum, providing indirect evidence for altered DA neu- AN relative to different conditions. Prior to eating a meal,
rotransmission in AN. AN participants have been found to show significantly
Neuropeptides have also been of interest in AN as they decreased rCBF in the left parietal cortex in comparison to
are involved in the regulation of feeding behaviours. controls. Post-meal, however, AN participants were found
Abnormalities in the ill state of AN have been found for a to not differ in rCBF relative to healthy controls or BN
variety of neuropeptides including opioid peptides, oxytocin, patients (Nozoe et al., 1995). Individuals with AN have also

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 16 31/10/2013 10:22:18 AM

 Phillipou et al. 17

been found to show differences in rCBF when viewing images of food and non-food items on the other hand, Joos
stimuli of their own bodies. In a study by Beato-Fernández et al. (2011) found increased right amygdala activity and
et al. (2009), healthy individuals, patients with BN and decreased posterior middle cingulate cortex activity in
patients with AN viewed videos of themselves and land- AN-R compared to controls, whereas Kim et al. (2012)
scapes. When viewing videos of one’s own body, increased reported increased left anterior insula activity in AN, and
rCBF of the left parietal and right superior frontal cortices significant interactions between the right insula and inferior
was found in AN, whereas healthy controls showed frontal gyrus. However, it must be noted that the studies by
decreased activation in these areas, suggesting a possible Kim et al. (2012) and Joos et al. (2011) involved passive
dysfunction in somatosensory integration in AN. viewing of images. In a study undertaken by Brooks et al.
(2011), the investigators asked participants to imagine they
were eating the food in the images presented to them and
Glucose metabolism were using the objects in the control images. Increased
Findings similar to those reported with the use of SPECT activity of the cerebellum, left visual cortex, right DLPFC
have also been found with PET. Individuals with AN have and parietal lobe was reported in AN-R to food compared to
been shown to display global hypometabolism as well as non-food items, and in the bilateral cerebellum and supple-
relative hypometabolism of glucose in cortical regions, mentary motor area in AN-BP. However, this study only
most significantly in the frontal and parietal cortices reported within-group comparisons for AN participants and
(Delvenne et al., 1995, 1997), which have been found to did not present between-group comparisons with healthy
normalise with weight gain (Delvenne et al., 1996). The controls. Utilising the same paradigm, a more recent study
same authors also reported hypermetabolism in the inferior by this group, which reported results compared to healthy
frontal cortex and basal ganglia when compared to con- individuals, increased right visual cortex activity and
trols, as well as increased glucose metabolism in the cau- reduced bilateral cerebellar vermis activity was reported in
date and putamen when compared to participants with BN AN when thinking about eating food (Brooks et al., 2012).
(Delvenne et al., 1999); whereas AN individuals in remis- When the investigators examined AN-R and AN-BP sepa-
sion have been shown to display normal glucose metabo- rately in comparison to healthy controls, increased right
lism in the brain (Delvenne et al., 1996; Frank et al., 2007; visual cortex and DLPFC activity, and reduced left cerebel-
Herholz et al., 1987). Furthermore, individuals with AN lar vermis and right insular activity was found when think-
have been shown to have greater regional cerebral blood ing about eating food in AN-R. Similar results were
flow within the bilateral medial temporal lobe in compari- reported for the AN-BP group, although no increase in right
son to healthy controls (Gordon et al., 2001), an area heav- DLPFC activity was found. When comparing the two AN
ily involved in memory processes (Squire and Zola-Morgan, subgroups to one another, the AN-R group showed
1991) and an area which has also been associated with increased activity of the visual cortex, left parahippocam-
increased blood flow in schizophrenia (Friston et al., 1992). pal gyrus and left ACC, suggesting a possible difference in
Additionally, the study by Gordon et al. (2001) reported the processing of food-related information between the two
increased occipital and temporo-occipital cortex activity of AN subgroups.
AN patients relative to controls when viewing images of BOLD activity differences in response to images of food
high- compared to low-calorie foods. These areas are have also been found between patients currently ill with
related to the processing of visual information and increased AN and weight-recovered patients. Holsen et al. (2012)
activity of related visual processing areas has been found in presented ill AN, weight-recovered AN and healthy con-
individuals with specific phobias when presented with pho- trols with images of food and non-food items before and
bic stimuli (Wik et al., 1993). after a meal. Pre-meal, the two patient groups showed
reduced hypothalamus, amygdala and anterior insula activ-
ity when viewing images of high-calorie foods compared to
Blood oxygen level dependent (BOLD) controls. Additionally, the ill AN group also showed
response increased hippocampus and orbitofrontal cortex activity.
Changes in neural activity in response to different condi- Post-meal, the group with active illness persisted to show
tions that were initially explored with PET have typically reduced amygdala and anterior insula activity, whereas the
been explored with functional magnetic resonance imaging weight-recovered individuals did not. Other studies assess-
(fMRI) since the technology became widely available. In ing AN participants in a hungry and satiated state have
the first fMRI study in AN, Ellison et al. (1998) presented reported increased posterior cingulate (Gizewski et al.,
individuals with images of high- and low-calorie drinks. 2010), and inferior parietal and visual cortex activity
The authors reported that the group of AN participants (Santel et al., 2006) in a hungry state when viewing food
showed increased activity of the left insula, ACC and images. When satiated, increased activity of the left insula
amygdala-hippocampal region in response to high- versus to food images has been reported in AN in comparison to
low-calorie drinks, relative to controls. In response to controls (Gizewski et al., 2010). In relation to a hungry

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 17 31/10/2013 10:22:18 AM

 18 ANZJP Articles

state, however, increased occipital cortex activity has been a current diagnosis of AN-R, the weight-recovered group
reported in AN when viewing food images (Santel et al., were found to show heightened activation of the dorsal
2006). Different neural responses in AN when hungry and ACC and the prefrontal cortex. In this earlier study, group
satiated have also been investigated with the use of oral differences were specific to food stimuli and no difference
stimuli. In a task where participants were required to drink in activation was apparent for the emotional aversive stim-
chocolate milk or water through a tube while undergoing uli for either AN-R or weight-recovered AN compared to
fMRI, the group of AN-R participants showed increased controls. In the later study, however, which contained a
right amygdala and left medial temporal gyrus activity rela- larger sample size, AN participants were found to show
tive to controls when drinking the flavoured milk (Vocks reduced right cerebellar activity but increased left cerebel-
et al., 2011). Although groups did not differ in responses to lar activity when viewing the emotional aversive images
the chocolate milk in the satiated condition, when compar- compared to healthy individuals. Individuals with AN have
ing the hungry to the satiated condition the AN group dis- also been found to show differences in brain activation
played increased inferior temporal gyrus activity, which when presented with negative words related to interper-
included the extrastriate body area, whereas the control sonal relationships. Specifically, AN has been associated
group showed increased left insula activity. In a study by with increased left insula, and right superior temporal and
Wagner et al. (2007a), on the other hand, where participants inferior frontal gyri activation, relative to healthy controls
were administered either a sucrose solution or water, the (Miyake et al., 2012). In an earlier study, the same group
investigators reported reduced activation to both stimulus investigated neural responses to negative body-image
types in the insula, and ventral and dorsal striatum in AN. A words (Miyake et al., 2010). AN-R was associated with
similar finding was reported by Oberndorfer et al. (2013), more activity of the right amygdala and inferior parietal
who reported reduced right insula activity in AN to both lobule, whereas AN-BP was associated with higher activity
sucrose and sucralose in a group of weight-recovered AN of the amygdala and left ventromedial prefrontal cortex,
participants relative to healthy controls. Furthermore, dif- relative to a healthy comparison group. Responses to nega-
ferences in striatum and insula activity have also been tive body-image words have also been investigated in AN,
reported in a study by Cowdrey et al. (2011). In this study utilising an Emotional Stroop task consisting of fat, thin
the investigators administered a pleasant chocolate taste or and neutral words, and words comprised of XXXXs
an aversive unpleasant taste coupled with the presentation (Redgrave et al., 2008). This study revealed increased acti-
of chocolate, mouldy strawberries or a grey control image, vation at the junction of the left insula, and frontal and tem-
or a grey screen coupled with a tasteless liquid which acted poral lobes, and the left middle and medial frontal gyri to
as a control condition, to healthy individuals and individu- thin versus XXXX words in AN compared to controls;
als weight-recovered from AN-R. The weight-recovered whereas the fat versus XXXX contrast revealed lower
AN group was found to show increased ventral striatum activity of the DLPFC and parietal areas in comparison to
and cingulate cortex activity to the taste of chocolate, and controls. These findings suggest that individuals with AN
increased activity of the cingulate, medial prefrontal and process positive and negative body image words differently
occipital cortices in response to the sight of chocolate. to healthy individuals, with increased activation to thin
When presented with the mouldy strawberry picture, how- words and decreased activation to fat words in areas related
ever, no group differences were apparent, although to executive function.
increased activation of a number of areas was evident when Furthermore, body image in AN has also been explored
the oral stimulus was administered alone or with the corre- by examining responses to images of human bodies. A
sponding aversive image, including the ACC, DLPFC, lat- recent study involving passive viewing of body images and
eral posterior insula, putamen and caudate. images of chairs, which acted as control condition, aimed
Furthermore, in a series of studies undertaken by Uher to investigate the connectivity within the core network for
and colleagues (Uher et al., 2003, 2004), the investigators body processing (Suchan et al., 2013). The findings of the
presented participants with images of food and non-food study revealed a different network in AN during the pro-
items and asked them to think about how hungry the images cessing of human bodies. In healthy controls, effective con-
made them; and emotional aversive and neutral images nectivity was found between the middle occipital gyrus and
which required participants to think about how the images extrastriate body area, and the extrastriate body area and
made them feel. In the food/non-food contrast, individuals fusiform gyrus. In AN, however, effective connectivity was
with AN demonstrated more activity of the lingual gyrus evident between the middle occipital gyrus and fusiform
and lower activity of the inferior parietal lobule and cere- body area, and the fusiform body area and the extrastriate
bellum when compared to controls (Uher et al., 2004); body area. Furthermore, reduced connectivity between the
whereas individuals weight-recovered from AN showed fusiform body area and extrastriate body area was found in
increased prefrontal and ACC activity, and reduced inferior the group of AN patients. Differences in the extrastriate
parietal activity in the same contrast compared to healthy body area and fusiform body area function between AN and
individuals (Uher et al., 2003). Relative to individuals with healthy individuals have also been reported in a study

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 18 31/10/2013 10:22:18 AM

 Phillipou et al. 19

involving the presentation of underweight, overweight and precuneus, posterior cingulate gyrus, fusiform gyrus, and
normal weight female bodies. Uher et al. (2005) reported right inferior and superior frontal gyri, parahippocampal
reduced activation of the occipitotemporal cortex (includ- gyrus, and bilateral inferior parietal lobule was reported in
ing the extrastriate body area), the fusiform gyrus and the AN (Vocks et al., 2010b). Furthermore, other studies have
parietal cortex in a group of AN participants to all body not only presented individuals with images of their own
shapes, relative to both BN and healthy controls. This task body, but also distorted images of their own body. Castellini
required participants to think about how acceptable the et al. (2013) asked participants to passively view images of
body or house (control image) would be for them. In a simi- their own body undistorted, their own body digitally dis-
lar study utilising stimuli of computer-generated nude torted, and control of images of houses. The authors
females depicting underweight, overweight and normal reported that in comparison to healthy controls, individuals
body weights, participants were asked to process the stim- with AN showed increased inferior frontal gyrus activity
uli during two conditions: one in which they were required when presented with overweight stimuli relative to normal
to imagine how it would feel to be the body shape pre- weight stimuli, and greater activity of the middle temporal
sented, and the other where they needed to estimate the gyrus in the underweight–normal weight contrast. In
body weight of the stimuli, which acted as the control con- another study with similarly distorted images, participants
dition (Fladung et al., 2010). Greater activity of the ventral were asked to either rate how much the image corresponded
striatum to normal weight compared to underweight stimuli to their real body, or how much it represented their ideal
was evident in healthy controls, whereas individuals with body (Mohr et al., 2010). In the ideal body condition,
AN were found to show greater activity to underweight increased activity of the insula and middle frontal gyrus
stimuli in the same contrast. However, groups were not was found for thin images in comparison to thin images of
found to differ in response to overweight stimuli. In another the real body condition. Additionally, the AN group were
study, Friederich et al. (2010) presented participants with found to show decreased precuneus activity for all body
images from magazines of either slim idealised female bod- sizes in the real body condition compared to controls, and
ies or interior home designs in which participants were to the heavier images in the real body condition relative to
asked to compare their body shape or room design with the the ideal body condition in comparison to healthy individu-
presented image. Group comparisons revealed that in als. Furthermore, an early fMRI study in AN undertaken by
response to the body images, individuals with AN showed Seeger et al. (2002) presented distorted own body images
decreased activity of the rostral ACC and greater activity of of maximum subjective unacceptability to the participant,
the right insula and premotor cortex. In a recent study by subjective unacceptability of another woman’s body and
Suda et al. (2013), the investigators presented individuals abstract images consisting of the participant’s own body.
with AN and healthy individuals with images of ‘body This study reported that AN participants showed increased
checking’, such as images of measuring leg width with a activity of the right amygdala, fusiform gyrus and brain-
measuring tape and pinching skin folds, and neutral active stem to distorted own body images compared to distorted
images such as using a computer or writing. The authors images of other bodies and the abstract images. However,
reported increased right parietal cortex activity in the AN this study was only comprised of three individuals with AN
group to the body checking images compared to the neutral and three comparison participants. In a follow-on study by
images, and lower activity of the anteromedial prefrontal the same group of investigators, the same task was per-
cortex and right fusiform gyrus relative to controls. formed in a larger sample and was found to lead to increased
Rather than displaying images of other people’s bodies, inferior parietal lobule and prefrontal cortex activity in the
several studies have focused on responses to images of same comparison mentioned above (Wagner et al., 2003).
one’s own body. Sachdev et al. (2008) found that in com- Although the great majority of fMRI studies have pre-
parison to a group of healthy participants, AN participants sented individuals with disorder-relevant stimuli such as
showed less activity of the frontal gyri, insula, precuneus bodies and food, other aspects of the disorder have also
and occipital regions when processing own body images, been investigated. A number of cognitive tasks have been
but did not differ when processing non-self body images. performed while undergoing fMRI, including tasks of inhi-
Vocks et al. (2010a), on the other hand, reported higher bition, working memory, visuospatial processing, cognitive
amygdala activity in AN compared to BN and controls flexibility and reward. During the completion of a stop-
when viewing another woman’s body, and reduced inferior signal task, groups of AN participants and healthy controls
parietal lobule activity when viewing own body images demonstrated similar neural activity when the inhibitory
compared to controls alone. In a related study, the same demand was low. However, when the inhibitory demand
authors presented AN patients with own body stimuli was high, individuals with AN showed decreased prefrontal
before and after undergoing body image therapy and found cortex activity in comparison to controls (Oberndorfer
increased pre- to post-treatment activity in the right middle et al., 2011). However, during a number n-back task, a task
temporal gyrus (including the extrastriate body area); assessing working memory, individuals with AN were
whereas decreased pre- to post-treatment activity in the left reported to show increased activation of temporal and

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 19 31/10/2013 10:22:18 AM

 20 ANZJP Articles

parietal areas compared to controls. A subgroup of the same combined. When a stimulus was unexpectedly deprived,
AN participants were retested following seven months of AN participants showed reduced activation of the bilateral
treatment and weight restoration and were found to show putamen and left orbitofrontal cortex compared to the two
reduced activity in the ACC, and frontal and parietal regions comparison groups. Additionally, in a study presenting a
compared to the results from the original test session. At simple monetary reward task to individuals weight-recov-
follow-up, the AN patients no longer differed from the con- ered from AN and a healthy comparison group, the healthy
trol group (Castro-Fornieles et al., 2010). However, in a controls were found to show increased left anterior ventral
more recent study by Lao-Kaim et al. (2013), the authors striatum activity for wins than for losses, whereas the
reported no difference in performance or brain activity patient group did not differ and had similar responses to
between groups of AN participants and healthy controls on both conditions, suggesting a possible difficulty in differen-
a letter n-back task, suggesting that verbal working mem- tiating positive and negative feedback in AN (Wagner et al.,
ory may be intact in AN. During a visuospatial processing 2007b).
task, on the other hand, in which participants were required A range of other BOLD studies have been undertaken in
to identify a target shape within more complicated figures, AN. One study aiming to assess emotion processing in AN
the AN group not only performed poorer on the task but presented individuals with happy and fearful human faces
also showed greater fusiform gyrus activity, whereas con- (Cowdrey et al., 2012b). The study revealed no difference
trols showed greater precuneus activity (Fonville et al., in neural response between groups. However, this result
2013). These findings suggest that unlike healthy individu- may be due to the fact that appropriate comparisons
als who utilise visuospatial search strategies to complete between conditions could not be made as a neutral control
such tasks, individuals with AN may utilise strategies condition was not presented. A study investigating the pro-
related to object recognition. During a behavioural response cessing of identity in AN, on the other hand, found reduced
shifting task, on the other hand, where participants were precuneus dorsal ACC and left middle frontal gyrus activity
required to respond to and classify shapes as targets, non- when presented with stimuli related to self-knowledge and
targets or standard shapes, Zastrow et al. (2009) described perspective taking (McAdams and Krawczyk, 2012). Two
that individuals with AN showed decreased bilateral thala- recent studies have also looked at pain perception in AN. In
mus, rostral cingulate, sensorimotor region and cerebellum the earlier study, Bär et al. (2012) administered a painful
activity to target trials, independent of performance. The thermal stimulus to the right arm of participants and found
AN group also showed reduced activity of the dorsal ACC that AN participants had an increased heat pain threshold
and ventral putamen in the correct shift target trials, and and increased right pons activity in comparison to controls.
reduced activity in the rostral ACC and dorsal and middle The later study undertaken by Strigo et al. (2013) investi-
striatum in the correct maintain target trials, relative to con- gated pain processing and anticipation in weight-recovered
trols. Incorrect trials relative to correct trials, on the other AN. The authors reported increased DLPFC and decreased
hand, were associated with lower activity of the dorsal posterior insula activity during pain stimulation in the AN
ACC, which extended to the medial frontal gyrus in AN. group. During pain anticipation, however, the AN group
Together, the findings by Zastrow et al. (2009) indicate that displayed greater right anterior insula, DLPFC and cingu-
during tasks of behavioural response shifting, individuals late cortex activity anticipation, relative to controls, sug-
with AN may have altered fronto-striatal-thalamic func- gesting a potentially intensified stress response.
tion. Additionally, during the Wisconsin Card Sort Test, a Recently, studies have begun to investigate functional
task of behavioural response shifting and cognitive flexibil- connectivity in individuals with AN. During resting state
ity, individuals with AN are also found to show differences scans of individuals weight-recovered from AN, increased
in neural response. In a study by Sato et al. (2013), the temporal coherence has been reported between the default
authors reported poorer performance on the task in AN, and mode network functional connectivity map and the right
lower right ventrolateral prefrontal cortex and bilateral par- precuneus and the DLPFC/inferior frontal gyrus (Cowdrey
ahippocampal cortex activity relative to healthy controls. et al., 2012a). Decreased connectivity in the right middle
Differences in responses to stimuli related to reward frontal gyrus has also been reported for weight-recovered
have also been recently investigated in AN. In a study by AN, and decreased connectivity of the left occipitotempo-
Frank et al. (2012), participants performed a reward-condi- ral junction within the ventral visual network (Favaro et al.,
tioning task where they not only learned specific associa- 2013). Favaro et al. (2013) also reported increased connec-
tions, but they would receive or be deprived of one of the tivity in the left superior parietal cortex within the soma-
two association stimuli, therefore causing an unexpected tosensory network for individuals with AN. In another
violation of these learned associations. When an unex- study, the same authors described increased prefrontal
pected stimulus was received, AN participants had greater functional connectivity in individuals with AN who were
right ventral putamen activation compared to controls, and Met-Met carriers of the COMT gene, compared to AN
greater left orbitofrontal cortex activity in comparison to a patients who were Val carriers of the gene (Favaro et al.,
sample of obese participants and the healthy control group 2012). In a more recent study by Amianto et al. (2013), the

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 20 31/10/2013 10:22:18 AM

 Phillipou et al. 21

investigators reported intrinsic connectivity networks in the 1986). It is also involved in regulating feeding behaviour
cerebellum in AN which were more restricted to vermian such as playing an inhibitory role in eating, regulating meal
areas, and showed greater medial and less lateral connectiv- size and controlling eating rate (Blundell, 1984; Leibowitz
ity in the cerebellum relative to healthy individuals. The and Alexander, 1998; Simansky, 1995).
AN group also demonstrated less cerebellar connectivity SPECT studies have revealed that individuals with AN
with the parietal lobe, but more cerebellar connectivity show hypoperfusion in several brain areas, although
with the posterior cingulate cortex, bilateral temporal pole whether these rCBF differences persist into weight recov-
and insula. Furthermore, the anterior insula showed more ery remains unclear. Similarly, PET studies have demon-
connectivity in AN, in contrast to individuals with BN who strated global hypometabolism in individuals ill with AN,
showed hyperconnectivity with the posterior portion of the but this change in glucose metabolism is no longer evident
insula. in weight-recovered individuals. Functional differences in
the brains of these individuals have however tended to be
Conclusion investigated with more sophisticated neuroimaging tech-
niques such as fMRI.
The objective of this review was to provide a summary of Recent fMRI studies have provided us with a better
the current literature on the neurobiology of AN. It is appar- understanding of the areas of the brain potentially responsi-
ent that structural and functional changes occur during the ble for the onset or maintenance of the illness. The function
ill state of AN, some of which have been found to reverse of the insular and cingulate cortices, in particular, have
with weight restoration, and others which persist following been regularly reported to differ in AN. The cingulate cor-
weight recovery. A number of structural brain differences tex is particularly involved in motivation, goal-directed
have been reported in AN. The enlargement of cortical sulci behaviour and emotional processes (Devinsky et al., 1995);
and ventricles is commonly reported, as are deficits in grey whereas the insula also plays a key role in emotional pro-
and white matter volumes. Although reduced grey matter cesses (Jones et al., 2010), particularly in the processing of
volumes are frequently found to persist in weight recovery, disgust (Wicker et al., 2003), an emotion which AN patients
the findings regarding white matter volumes in both weight- have been reported to have particular difficulty recognising
recovered and ill AN patients remain inconsistent. Recent (Pollatos et al., 2008).
DTI studies have suggested poorer white matter integrity in Due to inconsistencies across study procedures, findings
a number of brain areas in AN, although a limited number are not always consistent and can be difficult to interpret. A
of studies utilising this technique have thus far been under- major issue in all AN research is that it remains unclear
taken. Grey matter deficits in specific brain areas have also whether changes occur as a result of the condition and are
been reported, particularly areas of the limbic system due to the effects of starvation, or whether they are innate
including the amygdala, hippocampus and cingulate cortex, differences that contribute to the development of the ill-
areas heavily involved in emotion (LeDoux, 2000). ness. This is a very challenging issue to address as it would
Reduced grey matter volumes have also frequently been be difficult, if not impossible, to test individuals before the
reported in the putamen, a structure of the basal ganglia onset of illness symptoms. This is one of the reasons why
involved in learning, as well as the regulation of DA recovered AN patients are often researched. However, the
(Packard and Knowlton, 2002). findings observed at recovery may be state effects, reflect-
Deficits in DA function have been reported in AN, spe- ing damage due to starvation, rather than trait effects of
cifically, reduced levels of CSF HVA and altered DA recep- AN. Additionally, the length of time in which patients are
tor function. DA plays an important role in eating weight-recovered varies between studies as a standard min-
behaviours, motivation, reinforcing behaviour and reward imum timeframe for what constitutes someone who is in
(Bassareo and Di Chiara, 1999; Holroyd and Coles, 2002; recovery does not exist. Furthermore, patients with varying
Phillips et al., 2008; Roitman et al., 2004; Schultz et al., illness severity and duration are often included in the same
1997; Volkow et al., 2003; Wise, 2004); behaviours which analysis. Although particularly difficult to control due to
are also found to be disturbed in AN (e.g. Fladung et al., small patient numbers and often limited information regard-
2010; Scheurink et al., 2010; Wagner et al., 2007b; Watson ing illness duration, more systematic recruitment proce-
et al., 2010). Similarly, deficits in 5-HT function have also dures would provide a better understanding of how these
been reported in AN, namely, increased 5-HT1A and factors may contribute to neurobiological findings.
decreased 5-HT2A binding in ill and weight-recovered AN, Furthermore, if information is gathered about illness dura-
and decreased CSF 5-HIAA in ill patients but increased tion and severity, correlational analyses could be performed
levels in weight-recovered patients. 5-HT plays a role in a with neurobiological data.
number of symptoms and behaviours which are evident in The results of this review suggest a number of structural
the condition, including obsessional behaviours, anxiety, and functional deficits in the brains of individuals with AN as
impulse control, inhibition, attention and mood (Buhot, revealed with the use of a number of neuroimaging modalities.
1997; Higley and Linnoila, 1997; Lucki, 1998; Soubrie, It is important to note that some modalities were not included

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 21 31/10/2013 10:22:19 AM

 22 ANZJP Articles

in this review, such as electroencephalography (EEG), as they positron emission tomography and [carbonyl11C] WAY-100635.
were beyond the scope of this paper. However, a number of Archives of General Psychiatry 62: 1032.
Bailer UF and Kaye WH (2003) A review of neuropeptide and neuroen-
recent papers have suggested EEG and event-related potential docrine dysregulation in anorexia and bulimia nervosa. Current Drug
differences in AN (e.g. Hatch et al., 2010, 2011; for a review Targets. CNS and Neurological Disorders 2: 53–59.
see Jáuregui-Lobera, 2012), and further studies utilising more Bailer UF, Narendran R, Frankle WG, et al. (2012) Amphetamine induced
sophisticated neuroimaging techniques such as magnetoen- dopamine release increases anxiety in individuals recovered from ano-
cephalography will provide further information regarding dif- rexia nervosa. International Journal of Eating Disorders 45: 263–271.
Bailer UF, Price JC, Meltzer CC, et al. (2004) Altered 5-HT2A receptor bind-
ferences in the time course of activity in AN. ing after recovery from bulimia-type anorexia nervosa: relationships
The findings of this review have important implications to harm avoidance and drive for thinness. Neuropsychopharmacology
for the treatment of AN, however, more systematic research 29: 1143–1155.
with larger sample sizes is required. Of great benefit would Bär KJ, Berger S, Schwier C, et al. (2012) Insular dysfunction and
also be research which utilises tasks that are known to descending pain inhibition in anorexia nervosa. Acta Psychiatrica
Scandinavica 127: 269–278.
employ the areas which are disturbed in AN, such as emo- Bassareo V and Di Chiara G (1999) Differential responsiveness of dopa-
tion processing tasks which employ areas such as the insula mine transmission to food-stimuli in nucleus accumbens shell/core
and limbic system, as well as saccadic eye movement tasks compartments. Neuroscience 89: 637–641.
which involve the insula, cingulate cortex, basal ganglia, Beato-Fernández L, Rodríguez-Cano T, García-Vilches I, et al. (2009)
DLPFC and striatum, among other areas. With further Changes in regional cerebral blood flow after body image exposure in
eating disorders. Psychiatry Research: Neuroimaging 171: 129–137.
investigation into the neurobiology of AN, a better under- Blundell JE (1984) Serotonin and appetite. Neuropharmacology 23:
standing of the mechanisms involved in the development 1537–1551.
and maintenance of the illness will be established which Boghi A, Sterpone S, Sales S, et al. (2011) In vivo evidence of global
will have the potential to provide new direction for the and focal brain alterations in anorexia nervosa. Psychiatry Research:
development of effective treatments. Neuroimaging 192: 154–159.
Brambilla F, Bellodi L, Arancio C, et al. (2001) Central dopaminergic
function in anorexia and bulimia nervosa: a psychoneuroendocrine
Acknowledgements approach. Psychoneuroendocrinology 26: 393–409.
Brooks SJ, O’Daly O, Uher R, et al. (2012) Thinking about eating food
The authors would like to thank Larry Abel, PhD, for his contribu-
activates visual cortex with reduced bilateral cerebellar activation in
tion to the manuscript.
females with anorexia nervosa: An fMRI study. PLoS ONE 7: e34000.
Brooks SJ, Owen GO, Uher R, et al. (2011) Differential neural responses
Funding to food images in women with bulimia versus anorexia nervosa. PLoS
ONE 6: e22259.
This research was supported by a grant received from the Jack Buhot MC (1997) Serotonin receptors in cognitive behaviors. Current
Brockhoff Foundation (3410), a donation from the Dick and Pip Opinion in Neurobiology 7: 243–254.
Smith Foundation, and an Australian Postgraduate Award. Castellini G, Polito C, Bolognesi E, et al. (2013) Looking at my body.
Similarities and differences between anorexia nervosa patients and
controls in body image visual processing. European Psychiatry 28:
Declaration of interest
427–435.
The authors report no conflicts of interest. The authors alone are Castro-Fornieles J, Bargalló N, Lázaro L, et al. (2009) A cross-sectional
responsible for the content and writing of the paper. and follow-up voxel-based morphometric MRI study in adolescent
anorexia nervosa. Journal of Psychiatric Research 43: 331–340.
Castro-Fornieles J, Caldú X, Andrés-Perpiņá S, et al. (2010) A cross-sec-
References tional and follow-up functional MRI study with a working memory task
American Psychiatric Association (2013) Diagnostic and Statistical in adolescent anorexia nervosa. Neuropsychologia 48: 4111–4116.
Manual of Mental Disorders (DSM-5), Washington, DC: American Chowdhury U, Gordon I, Lask B, et al. (2003) Early-onset anorexia ner-
Psychiatric Association. vosa: Is there evidence of limbic system imbalance? International
Amianto F, D’Agata F, Lavagnino L, et al. (2013) Intrinsic connectiv- Journal of Eating Disorders 33: 388–396.
ity networks within cerebellum and beyond in eating disorders. Connan F, Murphy F, Connor SEJ, et al. (2006) Hippocampal volume
Cerebellum 12: 623–631. and cognitive function in anorexia nervosa. Psychiatry Research:
Artmann H, Grau H, Adelmann M, et al. (1985) Reversible and non- Neuroimaging 146: 117–125.
reversible enlargement of cerebrospinal fluid spaces in anorexia ner- Cowdrey FA, Filippini N, Park RJ, et al. (2012a) Increased resting state
vosa. Neuroradiology 27: 304–312. functional connectivity in the default mode network in recovered ano-
Audenaert K, Van Laere K, Dumont F, et al. (2003) Decreased 5-HT2a rexia nervosa. Human Brain Mapping. Epub ahead of print 3 October
receptor binding in patients with anorexia nervosa. Journal of Nuclear 2012. DOI: 10.1002/hbm.22202.
Medicine 44: 163–169. Cowdrey FA, Harmer CJ, Park RJ, et al. (2012b) Neural responses to emo-
Bailer UF, Frank GK, Henry SE, et al. (2007a) Serotonin transporter bind- tional faces in women recovered from anorexia nervosa. Psychiatry
ing after recovery from eating disorders. Psychopharmacology 195: Research 201: 190–195.
315–324. Cowdrey FA, Park RJ, Harmer CJ, et al. (2011) Increased neural process-
Bailer UF, Frank GK, Henry SE, et al. (2007b) Exaggerated 5-HT1A but ing of rewarding and aversive food stimuli in recovered anorexia ner-
normal 5-HT2A receptor activity in individuals ill with anorexia ner- vosa. Biological Psychiatry 70: 736–743.
vosa. Biological Psychiatry 61: 1090–1099. Delvenne V, Goldman S, Biver F, et al. (1997) Brain hypometabolism of
Bailer UF, Frank GK, Henry SE, et al. (2005) Altered brain serotonin 5-HT1A glucose in low-weight depressed patients and in anorectic patients: a
receptor binding after recovery from anorexia nervosa measured by consequence of starvation? Journal of Affective Disorders 44: 69–77.

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 22 31/10/2013 10:22:19 AM

 Phillipou et al. 23

Delvenne V, Goldman S, De Maertelaer V, et al. (1999) Brain glucose receptor binding in right frontotemporal cortex of both lean and recov-
metabolism in eating disorders assessed by positron emission tomog- ered patients: [18 F]MPPF PET scan study. Biological Psychiatry 64:
raphy. International Journal of Eating Disorders 25: 29–37. 1009–1013.
Delvenne V, Goldman S, De Maertelaer V, et al. (1996) Brain hypome- Gaudio S, Nocchi F, Franchin T, et al. (2011) Gray matter decrease dis-
tabolism of glucose in anorexia nervosa: normalization after weight tribution in the early stages of Anorexia Nervosa restrictive type in
gain. Biological Psychiatry 40: 761–768. adolescents. Psychiatry Research 191: 24–30.
Delvenne V, Lotstra F, Goldman S, et al. (1995) Brain hypometabolism of Giordano GD, Renzetti P, Parodi RC, et al. (2001) Volume measurement
glucose in anorexia nervosa: a PET scan study. Biological Psychiatry with magnetic resonance imaging of hippocampus-amygdala forma-
37: 161–169. tion in patients with anorexia nervosa. Journal of Endocrinological
Devinsky O, Morrell MJ and Vogt BA (1995) Contributions of anterior Investigation 24: 510–514.
cingulate cortex to behaviour. Brain 118: 279–306. Gizewski ER, Rosenberger C, de Greiff A, et al. (2010) Influence of sati-
Dolan RJ, Mitchell J and Wakeling A (1988) Structural brain changes in ety and subjective valence rating on cerebral activation patterns in
patients with anorexia nervosa. Psychological Medicine 18: 349–353. response to visual stimulation with high-calorie stimuli among restric-
Doraiswamy PM, Krishnan KR, Boyko OB, et al. (1991) Pituitary abnormal- tive anorectic and control women. Neuropsychobiology 62: 182–192.
ities in eating disorders: further evidence from MRI studies. Progress Golden NH, Ashtari M, Kohn MR, et al. (1996) Reversibility of cer-
in Neuro-Psychopharmacology and Biological Psychiatry 15: 351–356. ebral ventricular enlargement in anorexia nervosa, demonstrated by
Eckert ED, Halmi KA, Marchi P, et al. (1995) Ten-year follow-up of ano- quantitative magnetic resonance imaging. Journal of Pediatrics 128:
rexia nervosa: clinical course and outcome. Psychological Medicine 296–301.
25: 143–156. Gordon CM, Dougherty DD, Fischman AJ, et al. (2001) Neural substrates
Ellison Z, Foong J, Howard R, et al. (1998) Functional anatomy of calorie of anorexia nervosa: a behavioral challenge study with positron emis-
fear in anorexia nervosa. Lancet 352: 1192. sion tomography. Journal of Pediatrics 139: 51–57.
Favaro A, Clementi M, Manara R, et al. (2013) Catechol-O- Gordon I, Lask B, Bryant-Waugh R, et al. (1997) Childhood-onset anorexia
methyltransferase genotype modifies executive functioning and nervosa: Towards identifying a biological substrate. International
prefrontal functional connectivity in women with anorexia nervosa. Journal of Eating Disorders 22: 159–165.
Journal of Psychiatry & Neuroscience: JPN 38: 241–248. Harris EC and Barraclough B (1998) Excess mortality of mental disorder.
Favaro A, Santonastaso P, Manara R, et al. (2012) Disruption of visuos- British Journal of Psychiatry 173: 11–53.
patial and somatosensory functional connectivity in anorexia nervosa. Hatch A, Madden S, Kohn MR, et al. (2010) Emotion brain alterations in
Biological Psychiatry 72: 864–870. anorexia nervosa: a candidate biological marker and implications for
Fladung AK, Grön G, Grammer K, et al. (2010) A neural signature of ano- treatment. Journal of Psychiatry & Neuroscience: JPN 35: 267–274.
rexia nervosa in the ventral striatal reward system. American Journal Hatch A, Madden S, Kohn MR, et al. (2011) EEG in adolescent anorexia
of Psychiatry 167: 206–212. nervosa: impact of refeeding and weight gain. International Journal
Fonville L, Lao-Kaim NP, Giampietro V, et al. (2013) Evaluation of of Eating Disorders 44: 65–75.
enhanced attention to local detail in anorexia nervosa using the Herholz K, Krieg JC, Emrich HM, et al. (1987) Regional cerebral glu-
embedded figures test; an fMRI study. PLoS ONE 8: e63964. cose metabolism in anorexia nervosa measured by positron emission
Frampton I, Watkins B, Gordon I, et al. (2011) Do abnormalities in tomography. Biological Psychiatry 22: 43–51.
regional cerebral blood flow in anorexia nervosa resolve after weight Higley JD and Linnoila M (1997) Low central nervous system serotoner-
restoration? European Eating Disorders Review 19: 55–58. gic activity is traitlike and correlates with impulsive behavior. Annals
Frank GK, Bailer UF, Henry SE, et al. (2005) Increased dopamine D2/ of the New York Academy of Sciences 836: 39–56.
D3 receptor binding after recovery from anorexia nervosa measured Holroyd CB and Coles MGH (2002) The neural basis of human error
by positron emission tomography and [11C] raclopride. Biological processing: reinforcement learning, dopamine, and the error-related
Psychiatry 58: 908–912. negativity. Psychological Review 109: 679.
Frank GK, Bailer UF, Meltzer CC, et al. (2007) Regional cerebral blood Holsen LM, Lawson EA, Blum J, et al. (2012) Food motivation circuitry
flow after recovery from anorexia or bulimia nervosa. International hypoactivation related to hedonic and nonhedonic aspects of hun-
Journal of Eating Disorders 40: 488–492. ger and satiety in women with active anorexia nervosa and weight-
Frank GK, Kaye WH, Meltzer CC, et al. (2002) Reduced 5-HT2A receptor restored women with anorexia nervosa. Journal of Psychiatry &
binding after recovery from anorexia nervosa. Biological Psychiatry Neuroscience: JPN 37: 322–332.
52: 896–906. Jáuregui-Lobera I (2012) Electroencephalography in eating disorders.
Frank GK, Shott ME, Hagman JO, et al. (2013) Alterations in brain structures Neuropsychiatric Disease and Treatment 8: 1.
related to taste reward circuitry in ill and recovered anorexia nervosa and Jones CL, Ward J and Critchley HD (2010) The neuropsychological
in bulimia nervosa. American Journal of Psychiatry 170: 1152–1160. impact of insular cortex lesions. Journal of Neurology, Neurosurgery
Frank GKW, Reynolds JR, Shott ME, et al. (2012) Anorexia nervosa & Psychiatry 81: 611–618.
and obesity are associated with opposite brain reward response. Joos AAB, Saum B, van Elst LT, et al. (2011) Amygdala hyperreactivity
Neuropsychopharmacology 37: 2031–2046. in restrictive anorexia nervosa. Psychiatry Research: Neuroimaging
Friederich H, Brooks S, Uher R, et al. (2010) Neural correlates of body dis- 191: 189–195.
satisfaction in anorexia nervosa. Neuropsychologia 48: 2878–2885. Katzman DK, Lambe EK, Mikulis DJ, et al. (1996) Cerebral gray matter
Friederich H, Walther S, Bendszus M, et al. (2012) Grey matter abnor- and white matter volume deficits in adolescent girls with anorexia
malities within cortico-limbic-striatal circuits in acute and weight- nervosa. Journal of Pediatrics 129: 794–803.
restored anorexia nervosa patients. Neuroimage 59: 1106–1113. Katzman DK, Zipursky RB, Lambe EK, et al. (1997) A longitudinal mag-
Frieling H, Fischer J, Wilhelm J, et al. (2012) Microstructural abnormalities netic resonance imaging study of brain changes in adolescents with ano-
of the posterior thalamic radiation and the mediodorsal thalamic nuclei rexia nervosa. Archives of Pediatrics and Adolescent Medicine 151: 793.
in females with anorexia nervosa – a voxel based diffusion tensor imag- Kaye WH, Ebert MH, Gwirtsman HE, et al. (1984a) Differences in brain
ing (DTI) study. Journal of Psychiatric Research 46: 1237–1242. serotonergic metabolism between nonbulimic and bulimic patients with
Friston KJ, Liddle PF, Frith CD, et al. (1992) The left medial temporal anorexia nervosa. American Journal of Psychiatry 141: 1598–1601.
region and schizophrenia: a PET study. Brain 115: 367–382. Kaye WH, Ebert MH, Raleigh M, et al. (1984b) Abnormalities in CNS
Galusca B, Costes N, Zito NG, et al. (2008) Organic background of monoamine metabolism in anorexia nervosa. Archives of General
restrictive-type anorexia nervosa suggested by increased serotonin 1A Psychiatry 41: 350.

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 23 31/10/2013 10:22:19 AM

 24 ANZJP Articles

Kaye WH, Frank GKW and McConaha C (1999) Altered dopamine in anorexia nervosa. International Journal of Eating Disorders 41:
activity after recovery from restricting-type anorexia nervosa. 602–610.
Neuropsychopharmacology 21: 503–506. Miyake Y, Okamoto Y, Onoda K, et al. (2010) Neural processing of nega-
Kaye WH, Gwirtsman HE, George DT, et al. (1991) Altered seroto- tive word stimuli concerning body image in patients with eating dis-
nin activity in anorexia nervosa after long-term weight restoration: orders: An fMRI study. Neuroimage 50: 1333–1339.
Does elevated cerebrospinal fluid 5-hydroxyindoleacetic acid level Miyake Y, Okamoto Y, Onoda K, et al. (2012) Brain activation during
correlate with rigid and obsessive behavior? Archives of General the perception of stressful word stimuli concerning interpersonal rela-
Psychiatry 48: 556. tionships in anorexia nervosa patients with high degrees of alexithy-
Kaye WH, Gwirtsman HE, George DT, et al. (1988) CSF 5-HIAA concen- mia in an fMRI paradigm. Psychiatry Research: Neuroimaging 201:
trations in anorexia nervosa: Reduced values in underweight subjects 113–119.
normalize after weight gain. Biological Psychiatry 23: 102–105. Mohr HM, Zimmermann J, Röder C, et al. (2010) Separating two compo-
Kazlouski D, Rollin MDH, Tregellas J, et al. (2011) Altered fimbria-fornix nents of body image in anorexia nervosa using fMRI. Psychological
white matter integrity in anorexia nervosa predicts harm avoidance. Medicine 40: 1519–1529.
Psychiatry Research: Neuroimaging 192: 109–116. Mühlau M, Gaser C, Ilg R, et al. (2007) Gray matter decrease of the
Key A, O’Brien A, Gordon I, et al. (2006) Assessment of neurobiology anterior cingulate cortex in anorexia nervosa. American Journal of
in adults with anorexia nervosa. European Eating Disorders Review Psychiatry 164: 1850–1857.
14: 308–314. Naruo T, Nakabeppu Y, Deguchi D, et al. (2001) Decreases in blood per-
Kim KR, Ku J, Lee J-H, et al. (2012) Functional and effective connec- fusion of the anterior cingulate gyri in anorexia nervosa restricters
tivity of anterior insula in anorexia nervosa and bulimia nervosa. assessed by SPECT image analysis. BMC Psychiatry 1: 2.
Neuroscience Letters 521: 152–157. Naruo T, Nakabeppu Y, Sagiyama K, et al. (2000) Characteristic regional
Kingston K, Szmukler G, Andrewes D, et al. (1996) Neuropsychological cerebral blood flow patterns in anorexia nervosa patients with binge/
and structural brain changes in anorexia nervosa before and after purge behavior. American Journal of Psychiatry 157: 1520–1522.
refeeding. Psychological Medicine 26: 15–28. Neumärker KJ, Bzufka WM, Dudeck U, et al. (2000) Are there specific
Kojima S, Nagai N, Nakabeppu Y, et al. (2005) Comparison of regional disabilities of number processing in adolescent patients with anorexia
cerebral blood flow in patients with anorexia nervosa before and after nervosa? Evidence from clinical and neuropsychological data when
weight gain. Psychiatry Research: Neuroimaging 140: 251–258. compared to morphometric measures from magnetic resonance imag-
Komatsu H, Nagamitsu S, Ozono S, et al. (2010) Regional cerebral blood ing. European Child & Adolescent Psychiatry 9: 111–121.
flow changes in early-onset anorexia nervosa before and after weight Norring CEA and Sohlberg SS (1993) Outcome, recovery, relapse and
gain. Brain and Development 32: 625–630. mortality across six years in patients with clinical eating disorders.
Kuruoglu AC, Kapucu O, Atasever T, et al. (1985) Technetium-99m- Acta Psychiatrica Scandinavica 87: 437–444.
HMPAO brain SPECT in anorexia nervosa. Journal of Nuclear Nozoe SI, Naruo T, Yonekura R, et al. (1995) Comparison of regional
Medicine 39: 304–306. cerebral blood flow in patients with eating disorders. Brain Research
Lambe EK, Katzman DK, Mikulis DJ, et al. (1997) Cerebral gray mat- Bulletin 36: 251–255.
ter volume deficits after weight recovery from anorexia nervosa. Oberndorfer TA, Frank GKW, Simmons AN, et al. (2013) Altered insula
Archives of General Psychiatry 54: 537. response to sweet taste processing after recovery from anorexia and
Lao-Kaim NP, Giampietro VP, Williams SCR, et al. (2013) Functional MRI bulimia nervosa. American Journal of Psychiatry 170: 1143–1151.
investigation of verbal working memory in adults with anorexia nervosa. Oberndorfer TA, Kaye WH, Simmons AN, et al. (2011) Demand-specific
European Psychiatry (in press). doi 10.1016/j.eurpsy.2013.05.003. alteration of medial prefrontal cortex response during an inhibition
Lawrence AD, Dowson J, Foxall GL, et al. (2003) Impaired visual dis- task in recovered anorexic women. International Journal of Eating
crimination learning in anorexia nervosa. Appetite 40: 85–89. Disorders 44: 1–8.
Lázaro L, Andrés S, Calvo A, et al. (2013) Normal gray and white matter volume Packard MG and Knowlton BJ (2002) Learning and memory functions of
after weight restoration in adolescents with anorexia nervosa. International the basal ganglia. Annual Review of Neuroscience 25: 563–593.
Journal of Eating Disorders (in press). doi: 10.1002/eat.22161. Palazidou E, Robinson P and Lishman WA (1990) Neuroradiological and
LeDoux JE (2000) Emotion circuits in the brain. Annual Review of neuropsychological assessment in anorexia nervosa. Psychological
Neuroscience 23: 155–184. Medicine 20: 521–527.
Leibowitz SF and Alexander JT (1998) Hypothalamic serotonin in control Phillips AG, Vacca G and Ahn S (2008) A top-down perspective on dopa-
of eating behavior, meal size, and body weight. Biological Psychiatry mine, motivation and memory. Pharmacology Biochemistry and
44: 851–864. Behavior 90: 236–249.
Löwe B, Zipfel S, Buchholz C, et al. (2001) Long-term outcome of ano- Pollatos O, Herbert BM, Schandry R, et al. (2008) Impaired central
rexia nervosa in a prospective 21-year follow-up study. Psychological processing of emotional faces in anorexia nervosa. Psychosomatic
Medicine 31: 881–890. Medicine 70: 701–708.
Lucki I (1998) The spectrum of behaviors influenced by serotonin. Redgrave GW, Bakker A, Bello NT, et al. (2008) Differential brain activa-
Biological Psychiatry 44: 151–162. tion in anorexia nervosa to Fat and Thin words during a Stroop task.
Mainz V, Schulte-Rüther M, Fink GR, et al. (2012) Structural brain abnor- Neuroreport 19: 1181.
malities in adolescent anorexia nervosa before and after weight recov- Roberto CA, Mayer LES, Brickman AM, et al. (2011) Brain tissue vol-
ery and associated hormonal changes. Psychosomatic Medicine 74: ume changes following weight gain in adults with anorexia nervosa.
574–582. International Journal of Eating Disorders 44: 406–411.
Matsumoto R, Kitabayashi Y, Narumoto J, et al. (2006) Regional cer- Roitman MF, Stuber GD, Phillips PEM, et al. (2004) Dopamine operates
ebral blood flow changes associated with interoceptive awareness as a subsecond modulator of food seeking. Journal of Neuroscience
in the recovery process of anorexia nervosa. Progress in Neuro- 24: 1265–1271.
Psychopharmacology and Biological Psychiatry 30: 1265–1270. Sachdev P, Mondraty N, Wen W, et al. (2008) Brains of anorexia ner-
McAdams CJ and Krawczyk DC (2012) Who am I? How do I look? Neural vosa patients process self-images differently from non-self-images:
differences in self-identity in anorexia nervosa. Social Cognitive and an fMRI study. Neuropsychologia 46: 2161–2168.
Affective Neuroscience (in press) doi: 10.1093/scan/nss093 Santel S, Baving L, Krauel K, et al. (2006) Hunger and satiety in anorexia
McCormick LM, Keel PK, Brumm MC, et al. (2008) Implications of nervosa: fMRI during cognitive processing of food pictures. Brain
starvation-induced change in right dorsal anterior cingulate volume Research 1114: 138–148.

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd 24 31/10/2013 10:22:19 AM

 Phillipou et al. 25

Sato Y, Saito N, Utsumi A, et al. (2013) Neural basis of impaired cognitive Uher R, Murphy T, Friederich HC, et al. (2005) Functional neuroanatomy
flexibility in patients with anorexia nervosa. PLoS ONE 8: e61108. of body shape perception in healthy and eating-disordered women.
Scheurink AJW, Boersma GJ, Nergårdh R, et al. (2010) Neurobiology of Biological Psychiatry 58: 990–997.
hyperactivity and reward: Agreeable restlessness in anorexia nervosa. Vocks S, Busch M, Grönemeyer D, et al. (2010a) Neural correlates of
Physiology & Behavior 100: 490–495. viewing photographs of one’s own body and another woman’s body in
Schultz W, Dayan P and Montague PR (1997) A neural substrate of predic- anorexia and bulimia nervosa: an fMRI study. Journal of Psychiatry
tion and reward. Science 275: 1593–1599. & Neuroscience: JPN 35: 163.
Seeger G, Braus DF, Ruf M, et al. (2002) Body image distortion reveals Vocks S, Busch M, Schulte D, et al. (2010b) Effects of body image therapy
amygdala activation in patients with anorexia nervosa – a functional on the activation of the extrastriate body area in anorexia nervosa: An
magnetic resonance imaging study. Neuroscience Letters 326: 25–28. fMRI study. Psychiatry Research: Neuroimaging 183: 114–118.
Simansky KJ (1995) Serotonergic control of the organization of feeding Vocks S, Herpertz S, Rosenberger C, et al. (2011) Effects of gustatory
and satiety. Behavioural Brain Research 73: 37–42. stimulation on brain activity during hunger and satiety in females
Soubrie P (1986) Reconciling the role of central serotonin neurons in human with restricting-type anorexia nervosa: an fMRI study. Journal of
and animal behavior. Behavioral and Brain Sciences 9: 319–335. Psychiatric Research 45: 395–403.
Squire LR and Zola-Morgan S (1991) The medial temporal lobe memory Volkow ND, Wang GJ, Maynard L, et al. (2003) Brain dopamine is associ-
system. Science 253: 1380–1386. ated with eating behaviors in humans. International Journal of Eating
Strigo IA, Matthews SC, Simmons AN, et al. (2013) Altered insula acti- Disorders 33: 136–142.
vation during pain anticipation in individuals recovered from ano- Wagner A, Aizenstein H, Mazurkewicz L, et al. (2007a) Altered insula
rexia nervosa: Evidence of interoceptive dysregulation. International response to taste stimuli in individuals recovered from restricting-
Journal of Eating Disorders 46: 23–33. type anorexia nervosa. Neuropsychopharmacology 33: 513–523.
Strober M, Freeman R and Morrell W (1997) The long-term course of Wagner A, Aizenstein H, Venkatraman V, et al. (2007b) Altered reward
severe anorexia nervosa in adolescents: Survival analysis of recovery, processing in women recovered from anorexia nervosa. American
relapse, and outcome predictors over 10–15 years in a prospective Journal of Psychiatry 164: 1842–1849.
study. International Journal of Eating Disorders 22: 339–360. Wagner A, Ruf M, Braus DF, et al. (2003) Neuronal activity changes and
Suchan B, Bauser DS, Busch M, et al. (2013) Reduced connectiv- body image distortion in anorexia nervosa. Neuroreport 14: 2193.
ity between the left fusiform body area and the extrastriate body Watson KK, Werling DM, Zucker NL, et al. (2010) Altered social reward and
area in anorexia nervosa is associated with body image distortion. attention in anorexia nervosa. Frontiers in Psychology 1: 36.
Behavioural Brain Research 241: 80–85. Wicker B, Keysers C, Plailly J, et al. (2003) Both of us disgusted in
Suchan B, Busch M, Schulte D, et al. (2010) Reduction of gray matter my insula: The common neural basis of seeing and feeling disgust.
density in the extrastriate body area in women with anorexia nervosa. Neuron 40: 655–664.
Behavioural Brain Research 206: 63–67. Wik G, Fredrikson M, Ericson K, et al. (1993) A functional cerebral
Suda M, Brooks SJ, Giampietro V, et al. (2013) Functional neuroanatomy response to frightening visual stimulation. Psychiatry Research:
of body checking in people with anorexia nervosa. International Neuroimaging 50: 15–24.
Journal of Eating Disorders (in press). doi: 10.1002/eat.22150. Wise RA (2004) Dopamine, learning and motivation. Nature Reviews
Sullivan PF (1995) Mortality in anorexia nervosa. American Journal of Neuroscience 5: 483–494.
Psychiatry 152: 1073–1074. Yau WW, Bischoff-Grethe A, Theilmann RJ, et al. (2013) Alterations
Swayze VW, Andersen A, Arndt S, et al. (1996) Reversibility of brain tis- in white matter microstructure in women recovered from anorexia
sue loss in anorexia nervosa assessed with a computerized Talairach nervosa. International Journal of Eating Disorders (in press). doi:
3-D proportional grid. Psychological Medicine 26: 381–390. 10.1002/eat.22154.
Takano A, Shiga T, Kitagawa N, et al. (2001) Abnormal neuronal net- Yonezawa H, Otagaki Y, Miyake Y, et al. (2008) No differences are seen
work in anorexia nervosa studied with I-123-IMP SPECT. Psychiatry in the regional cerebral blood flow in the restricting type of anorexia
Research: Neuroimaging 107: 45–50. nervosa compared with the binge eating/purging type. Psychiatry and
Uher R, Brammer MJ, Murphy T, et al. (2003) Recovery and chronicity in Clinical Neurosciences 62: 26–33.
anorexia nervosa: brain activity associated with differential outcomes. Zastrow A, Kaiser S, Stippich C, et al. (2009) Neural correlates of impaired
Biological Psychiatry 54: 934–942. cognitive-behavioral flexibility in anorexia nervosa. American
Uher R, Murphy T, Brammer MJ, et al. (2004) Medial prefrontal cortex Journal of Psychiatry 166: 608–616.
activity associated with symptom provocation in eating disorders. Zipfel S, Löwe B, Reas DL, et al. (2000) Long-term prognosis in anorexia
American Journal of Psychiatry 161: 1238–1246. nervosa: lessons from a 21-year follow-up study. Lancet 355: 721–722.

Australian & New Zealand Journal of Psychiatry, 00(0)

Downloaded from anp.sagepub.com at Monash University on November 10, 2013

ANP509693.indd
View publication 25
stats 31/10/2013 10:22:19 AM