Systematic Review of fMRI Studies with Visual Food Stimuli in Anorexia Nervosa

Research Article

Austin J Nutr Metab. 2021; 8(4): 1115.

Systematic Review of fMRI Studies with Visual Food Stimuli in Anorexia Nervosa

Dabkowska-Mika A1,2, Steiger R1,2*, Gander M3, Sevecke K3 and Gizewski ER1,2

¹Department of Neuroradiology, Medical University of Innsbruck, Innsbruck, Austria

²Neuroimaging Research Core Facility, Innsbruck, Austria

³Department of Child and Adolescent Psychiatry, Medical University of Innsbruck, Innsbruck, Austria

*Corresponding author: Steiger R, Department of Neuroradiology, Medical University of Innsbruck, Anichstrasse 35, Innsbruck, Austria; Neuroimaging Research Core Facility, Innsbruck, Austria

Received: October 05, 2021; Accepted: November 08, 2021; Published: November 15, 2021

Abstract

Anorexia Nervosa (AN) is a disease with increasing prevalence and relatively high mortality that usually begins in adolescence. Patients with AN avoid food intake and may react specifically toward food images. We present a systematic review of fMRI studies with visual food stimulation in AN, based on a search through PubMed database under the recommendations of the PRISMA guidelines. After applying dates 2004.01.01-2021.01.01, we screened 319 papers and included 27 experimental designs, with only 7 studies focusing on adolescents. Adolescents with AN showed increased activity in the medial prefrontal cortex, the inferior frontal gyrus, the insula, the hippocampus, the fusiform gyrus, the parahippocampal gyrus and the cuneus when watching food images. Adult participants with AN revealed enhanced brain activity due to visual food stimuli in the fusiform gyrus, the inferior frontal gyrus, the lingual gyrus, the medial prefrontal cortex, the right dorsolateral prefrontal cortex, the right angular gyrus. There was deactivation detected in the parahippocampal gyrus, compared to healthy participants. We have found contrary reports according increased/decreased activation of the insula, the amygdala, the hippocampus, the hypothalamus, the anterior cingulate cortex, the thalamus, the orbitofrontal cortex in adults with AN.

Although AN typically develops in adolescence, there is still very little fMRI research in this age group. Careful creation of a homogeneous group of study participants is an important factor determining the reliability and unequivocalness of the experiment. Only a detailed description of participants´ characteristics that may affect the results allows solid comparison of different studies´ findings.

Keywords: Anorexia nervosa; Functional magnetic resonance imaging; Visual food stimuli; Adolescent psychiatry

Abbreviations

fMRI: Functional Magnetic Resonance Imaging; AN: Anorexia Nervosa; PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses; ED: Eating Disorder; BOLD: Blood Oxygenation Level Dependent; HC: Healthy Controls

Introduction

Anorexia Nervosa (AN) is an Eating Disorder (ED), characterised by restriction of food intake leading to significantly low body weight, intense fear of gaining weight and a distorted body image [1]. Although typically onset of AN is in adolescence [2], studies in this age group are relatively rare. Even though its prevalence rate is growing, it is still underdiagnosed [3]. Onset of AN often overlaps with increased vulnerability due to peers´ and social pressure, but also physical transmission from safe childhood into demanding adulthood. Juvenility is period of elevated need for calorie intake and last possibility to develop healthy body, with proper growth and brain volumes (brain consists in 60% of fat [4]). Limitation of calorie intake in juvenescence often results in significantly lower adult height [5,6]. Starvation and dehydration lead to brain volume loss [7] and influence cognitive processes. This can be crucial in adolescence, because it is usually time of attending final level of education and making decision about future life. As AN has the highest mortality rate of any psychiatric illness [8-10], it seems essential to understand both psychological and neural alterations underlying AN. Especially, that early age of onset, as well as short duration of symptoms and inpatient treatment are related to better prognosis [11].

Development of neuroimaging techniques aroused hope for quicker and more precise diagnose, for possibility to predict course of illness, and to find neuroimaging biomarkers. Although, the first paper performing neuroimaging in psychiatry concentrated on schizophrenia [12], it was soon followed by publication about adolescent anorectic patients [13]. However, among the 100 most highly cited papers about neuroimaging in psychiatry [14], there was no article about ED.

Functional Magnetic Resonance Imaging (fMRI) records activity of specific brain regions in vivo using the indirect detection of neuronal activity via hemodynamic changes. When activated, the brain area is supplied by a greater amount of oxygenated blood, so the ratio of oxygenated/deoxygenated haemoglobin is changed in vein vessels. Due to different magnetic properties, they can serve as intrinsic contrast agents and be detected by MR scanners. This method of imaging is relying on the BOLD (Blood Oxygenation Level Dependent) effect [15]. In order to analyse changed brain activation in a given disorder, one can use symptom-provoking paradigms. In AN such a disorder related stimuli can be, beside pictures of body shapes, food images. They are described as aversive, causing anxiety, even influencing cognitive performance, so they are triggers to cause specific for AN brain reaction, in comparison to Healthy Controls (HC) [16]. It was documented, that adolescents with AN respond faster to high-calorie food images than healthy participants [17].

We present a systematic review of papers related to fMRI studies employing experimental designs in AN using visual stimulation with images of food. Specifically, we focused on adolescents, as not many fMRI studies examined neural responses associated with AN in minors.

Material and Methods

To find matching articles, we have searched via PubMed, applying dates 2004.01.01 to 2021.01.01. The search strategy is presented in a Table 1.

Table 1: Systematic review search strategy.





  

    
Anorexia OR anorectic

    
AND

    
Image

      OR

      Imaging

      OR

      fMR*

      OR

      “Functional Magnetic Resonance Imaging”

      OR

      “Neural processing”

      OR

      Processing

    
AND

    
Visual OR Picture* OR Image OR Imaging

    
AND

    
Food OR Meal

  










Table 1: Systematic review search strategy.

We found 319 matching papers, then screening titles and abstracts we limited results to English language and original papers. Moreover, we excluded case reports, reviews and comorbidity papers. Furthermore, we eliminated studies concerning non-AN patients and animals. Additionally, we have searched through reference lists and eating disorders specialised journals. We were particularly interested in studies on adolescents.

Results

We screened 319 papers and finally included 27 in this review. We excluded 292 papers because of the mentioned exclusion criteria. Figure 1 shows PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) diagram (a tool suggested by Moher D with colleagues for systematic reviews) [18] (Figure 1).

Figure 1: Flowchart PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) presenting the main search strategy and article selection for systematic review.

    

    
    

    

    Figure 1: Flowchart PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) presenting the main search strategy and article selection for systematic review.

A summary of the results is shown in a Table 2.

Table 2: Characteristics of included studies.





  

    
Author

    
Participants

    
Viewed    images

    
Comments

    
Key    findings

  

  

    
Stimulus    and comparison

    
Results of    brain area

  

  

    
Horster et    al. [19]

    
31AN/27HC

    
Images of    food and objects

    
EDI-2,    BDI, EDE, MWT-B, STAI, rating of pictures after scanning.

      
Replication    study to one conducted by Joos AA et al., 2011

    
HC:    increased activation due to food stimuli:

    
Right    calcarine fissure, right middle occipital gyrus, left superior frontal gyrus,    left superior occipital gyrus, left insula, left superior parietal gyrus

  

  

    
AN:    increased activation due to food stimuli:

    
Left    middle occipital gyrus, right calcarine fissure, right lingual gyrus,    bilateral fusiform gyrus, left SMA, bilateral superior frontal gyrus, ACC,    left middle frontal gyrus (orbital part), left precuneus, bilateral insula,    bilateral midcingulate, right supramarginal gyrus, left postcentral gyrus,    right angular gyrus

  

  

    
AN vs. HC: increased activation due to    food stimuli:

    
Left MCC,    left precentral gyrus, left postcentral gyrus, left middle frontal cortex,    right IPL, right angular gyrus, right precuneus, right posterior cingulate    gyrus

  

  

    
HC vs. AN: activation due to food    stimuli:

    
No    significant results

  

  

    
Young    et al. [20] 

    
16AN/21HC

    
Images    of food (H and L) and objects 

    
SCID,    EDE-Q, YBC-EDS, DASS, PANAS, rating anxiety during fMRI task. 2 fMRI scans-    one before and one after 10 sessions of exposure-based therapy. 

    
AN vs. HC, pre-therapy, decreased    activation due to food stimuli:

    
ACC

  

  

    
AN vs. HC, post-therapy, increased    activation due to food stimuli:

    
DLPFC

  

  

    
AN vs. HC, post-therapy, decreased    activation due to food stimuli:

    
superior    parietal lobe

  

  

    
Association    between anxiety and changes of brain activation:

    
insula,    middle temporal gyrus/lateral parietal cortex

  

  

    
Ziv et    al. [21] 

    
11AN, 7    atypical AN

      13-18 yo.

    
Images of    food (sweet & non-sweet) and non-food 

    
EAT-26,    STAI 

    
increased    activation due to food vs. non-food    stimuli: 

    
Occipital    regions 

  

  

    
decreased    activation due to food vs. non-food    stimuli: 

    
Temporal    and parietal gyri 

  

  

    
increased    activation due to sweets versus non-food stimuli: 

    
Hippocampus 

  

  

    
increased    activation due to sweet vs. nonsweet food stimuli: 

    
OFC,    ACC 

  

  

    
Positive    correlation between STAI and brain activity, when comparing all foods versus    non-food stimuli: 

    
OFC,    ACC 

  

  

    
Positive    correlation between EAT-26 and brain activity, when comparing sweet versus    nonsweet stimuli: 

    
ACC,    frontal regions 

  

  

    
Stopyra    et al. [22] 

    
25AN/25HC

    
H and    non-food images 

    
Viewing    pictures or solving an arithmetic equation (distraction conditions). Infusion    of glucose/water through the nasogastric tube. 

      SCID,    EDE-Q, BDI, hunger rating, cravings rating. 

    
AN: due    to food vs. non-food distraction:    water comparing to glucose 

    
Precuneus

  

  
 

  

    
In a    state of hunger: AN vs. HC,    increased activation due to H vs. non-food distraction: 

    
Left    middle occipital gyrus, left inferior parietal lobule, left precuneus, left    fusiform gyrus 

  

  

    
In a    state of satiety: HC vs. AN,    increased activation due to H vs. non-food distraction: 

    
Left    PCC, left parahippocampal gyrus, left superior frontal gyrus, left medial    OFC, left ACC 

  

  

    
Negative    association between cravings rating in AN and brain activation: 

    
Bilateral    dorsal striatum 

  

  

    
Wein-bach    et al. [17] 

    
30AN/30HC

      12-18 yo.

    
H &    L 

    
SCID,    WASI-II, ED-Q, BDI, STAI, OCI. Food-stop signal task (response inhibition)    after food images presentation. 

    
Response    time to displayed images, not the specific brain region´s activation due to    stimuli. 

    
  

  

    
AN vs. HC- faster response to H 

  

  

    
Olivo    et al. [23] 

    
28atypicalAN/33HC

      13-16 yo.

    
H &    L 

    
EDE-Q,    MADRS, MINI-KID 

    
Functional    connectivity analysis not included in this table. 

    
  

  

    
Boehm    et al. [24] 

    
35AN/35HC 

      12-19    yo. 

    
30    neutral (i.e. house) and 30 happy social stimuli (i.e. children playing), as    well as food pictures (H & L, but not divided) 

    
Pictures    presented supraliminally and subliminally 

      SCID,    SIAB, EDI-2, WAIS, WISC, hunger rating. 

    
Supraliminal    stimuli:

    
Inferior    frontal junction (IFJ)

  

  

    
AN    increased activation due to all stimuli:

  

  

    
Supraliminal    stimuli: AN increased activation due to food stimuli:

    
Visual    regions (including superior occipital gyrus and the fusiform gyrus/    parahippocampal gyrus)

  

  

    
Subliminal    stimulation

    
No group    differences

  

  

    
Horn-dasch    et al. [25]

    
Adole-scents: 15AN/18HC 

      12-18    yo 

      Adults: 16AN/16HC 

      19-40    yo 

    
12    pictures of H & L food and 24 affective stimuli (IAPS) 

    
EDI-2,    BDI 

    
Adolescent AN vs. HC

  

  

    
AN    increased activation due to H stimuli:

    
IFG,    medial prefrontal gyrus, anterior insula

  

  

    
AN    decreased activation due to H simuli:

    
right    cerebellum

  

  

    
AN    increased activation due to L stimuli:

    
medial    prefrontal gyrus and inferior parietal cortex, cerebellum

  

  

    
AN    decreased activation due to L stimuli:

    
cerebellum

  

  

    
Adult AN vs. HC

  

  

    
AN    increased activation due to H & L stimuli:

    
cerebellum

  

  

    
AN    decreased activation due to L stimuli:

    
right    inferior frontal gyrus and thalamus

  

  

    
Adult AN vs. Adolescent AN

  

  

    
Adult AN    increased activation due to H stimuli:

    
superior    parietal and cerebellum

  

  

    
Adult AN    decreased activation due to L stimuli:

    
bilateral    superior frontal lobe, bilateral cingulate and left cerebellum

  

  

    
Adult HC vs. Adolescent HC 

  

  

    
Adult HC    increased activation due to H & L stimuli:

    
left    cerebellum

  

  

    
Adult HC    decreased activation due to H stimuli:

    
cingulate    cortex, insula and several cerebellar regions

  

  

    
Adult HC    decreased activation due to L stimuli:

    
caudate,    superior frontal gyrus and similar cerebellar regions

  

  

    
Kerr et    al. [26] 

    
20    weight restAN/ 

      20HC 

      13-24    yo 

    
Pictures    of high/ low palatability food, and objects 

    
Rating    of interoceptive sensations intensity, then comparing it with fMRI results. 

      SIAB-    EX, SCID, EDI, HAM-A. 

    
Relationship between stomach sensation intensity    ratings and brain activation

  

  

    
wrAN:    positive relationship due to high palatability stimuli:

    
amygdala    and subgenual ACC

  

  

    
HC:    negative relationship due to high palatability stimuli:

    
amygdala    and subgenual ACC

  

  

    
wrAN:    negative relationship due to low palatability stimuli:

    
ventral    pallidum, ventral tegmental area

  

  

    
HC:    positive relationship due to low palatability stimuli:

    
ventral    pallidum, ventral tegmental area

  

  

    
Scaife et    al. [27]

    
12AN/14    recAN/16HC 

    
40 H    & 40 L calorie food pictures 

    
EDE,    NART, STAI, BDI, YBC-EDS-SRQ, LOFPQ 

    
AN:    decreased activation due to food stimuli:

    
right    postcentral gyrus-precuneus, (extending to PCC), the left superior parietal    lobule-postcentral gyrus

  

  

    
RecAN vs. HC: activation due to food    stimuli:

    
no    significant differences

  

  

    
AN:    increased activation due to H stimuli:

    
right    lateral frontal pole

  

  

    
AN:    decreased activation due to L stimuli:

    
right    lateral frontal pole (also DLPFC), supramarginal gyrus

  

  

    
AN vs. recAN: activation due to L    stimuli:

    
no    significant differences

  

  

    
recAN vs. HC: activation due to L stimuli:

    
no    significant differences

  

  

    
Relationship between YBC-EDS-SRQ score ratings and    brain activation

  

  

    
AN:    negative relationship due to L stimuli:

    
frontal    pole

  

  

    
Connectivity analyses -Psychophysiological interaction

  

  

    
AN:    reduced coherence due to food stimuli between regions:

    
left    amygdala with caudate/putamen (dorsal striatum), dorsal ACC, medial PFC

  

  

    
the right    caudate with left postcentral gyrus – juxtapositional lobule cortex

  

  

    
AN:    reduced coherence due to H stimuli between regions:

    
left    caudate with the bilateral intracalcarine-lingual gyri

  

  

    
Sultson et    al. [28]

    
14 AN/ 

      14recAN    /15HC 

    
Images    of food (H and L) and objects 

    
Instruction    to imagine eating/ using what is presented. Rating anxiety and desire to eat. 

      STAI,    Beck, BCST 

    
Significant correlations between perseverative errors    and brain activation:

  

  
 

  

    
AN:    negative correlation during food processing:

    
dACC,    paracentral lobule, precuneus

  

  

    
AN:    negative correlation during non-food processing:

    
right    DLPFC

  

  

    
recAN:    positive correlation during food processing:

    
left dAAC    and VLPFC

  

  

    
recAN:    positive correlation during non-food processing:

    
left dAAC,    anterior insula and medial PFC

  

  

    
Significant correlations between non-perseverative    errors and brain activation:

  

  

    
AN:    negative correlation during food processing:

    
right dACC

  

  

    
HC:    positive correlation during non-food processing:

    
right dAAC    and DLPFC

  

  

    
Correlation between anxiety and brain activation:

  

  

    
AN:    negative correlation due to non-food stimuli:

    
precuneus

  

  

    
HC:    positive correlation due to food stimuli:

    
left DLPFC    and dACC

  

  

    
HC:    positive correlation due to non-food stimuli:

    
left DLPFC

  

  

    
Sanders et    al. [29]

    
15AN/ 

      14recAN/15HC 

    
Images    of food (H and L) and objects 

    
Instruction    to imagine eating/ using what is presented. 

      EDE-Q,    STAI 

    
AN:    increased activation due to food stimuli:

    
left    hippocampus, vermis, right cerebellum, hypothalamus, right middle frontal    gyrus, left inferior parietal cortex

  










Table 2: Characteristics of included studies.

Table 2 off 1: 





  

    
Holsen et al. [30] 

    
13AN/ 9wrAN /12HC 

    
Images of food (H and L) and    objects 

    
Results of neural activation only    in relation-ship with fasting plasma acylated ghrelin levels. 

      BDI, EDEQ, appetite ratings. 

    
Correlation    between fasting acylated ghrelin and brain activation: 

  

  

    
AN:    positive correlation due to L stimuli:

    
right OFC 

  

  

    
wrAN:    negative correlation due to H stimuli:

    
left hippocampus 

  

  

    
HC:    significant positive correlation due to H stimuli:

    
right    amygdala, hippocampus, insula, OFC

  

  

    
Relationship between desire to eat    and brain activation:

  

  

    
HC:    positive correlation due to H stimuli:

    
anterior insula 

  

  

    
HC:    positive correlation due to L stimuli:

    
OFC 

  

  

    
wrAN:    positive correlation due to L stimuli:

    
OFC 

  

  

    
Kull-mann et al. [31] 

    
12AN/12 athle-tes/14HC 

    
Images of food and objects; active    and non-active person. 

    
EDI-2, EDEQ, STAI, PHQ-D, BAS/BIS,    CES, hunger ratings. 

      Go/no-go tasks on every pairs of    pictures: 

      go food/ no-go object and go    object/ no go food; 

      go active /no-go inactive and go    inactive /no-go active. 

    
Response    inhibition across subjects to food vs. objects stimuli: 

  

  

    
Increased activation: 

    
right    middle and inferior temporal cortex, bilateral middle frontal cortex, right    supplementary motor area, left fusiform gyrus, bilateral insula, right    postcentral gyrus, left superior medial frontal gyrus, left supramarginal    gyrus, right superior frontal gyrus

  

  

    
Response inhibition to food vs. objects stimuli:

  

  

    
Increased activation: 

    
bilateral    fusiform gyrus and insula, inferior parietal gyrus and middle frontal gyrus

  

  

    
Decreased activation: 

    
mOFC,    middle temporal gyrus, PCC

  

  

    
Response inhibition to food vs. objects stimuli:

  

  

    
AN vs. HC: decreased activation:

    
right putamen 

  

  

    
AN vs. athletes: decreased activation:

    
right putamen 

  

  

    
AN vs. athletes and HC: reduced    response inhibition for food and objects stimuli.

  

  

    
Correlation between brain    activation during response inhibition for food/non-food stimuli and tests:

  

  

    
significant    negative correlation between EDI-2 score:

    
putamen 

  

  

    
positive    correlation between correct go responses:

    
putamen 

  

  

    
Results    of response inhibition to physical activity stimuli, as well as behavioural    results are not included in this review.

  

  

    
Lawson et al. [32] 

    
13AN/10wrAN/13HC 

    
Images of food (H and L) and    objects 

    
Comparing fMRI results with    peripheral cortisol and ACTH levels; in state of hunger and satiety. 

      BDI, SCID, appetite rating. 

    
Association between cortisol and ACTH levels and brain    activation due to H stimuli:

  

  

    
AN vs. HC: (premeal) increased    activation:

    
amygdala,    hippocampus, insula, hypothalamus, OFC

  

  

    
wrAN vs. HC: (premeal) increased    activation:

    
amygdala, insula, hypothalamus 

  

  

    
AN vs. HC: (postmeal) decreased    activation:

    
amygdala and insula 

  

  

    
AN vs. wrAN: (postmeal) increased    activation:

    
amygdala 

  

  

    
AN vs. wrAN: (postmeal) decreased    activation:

    
insula 

  

  

    
Obern-dorfer et al. [33] 

    
14recAN/12HC 

    
Images of food and objects 

    
Visual anticipatory task- a square    would be followed by food image and a circle by object image. Brain reaction    during watching food/object pictures, as well as during anticipation of    food/object pictures. 

      STAI, BDI, FMPS, TCI, BIS-11,    TAS-20. Pleasant-ness rating of pictures. 

      Scanning after meal. 

    
recAN during object anticipation: 

    
amygdala,    IFC, occipital lobes, anterior and superior cingulate gyrus

  

  

    
recAN during food anticipation: 

    
right    middle frontal gyrus, occipital lobes, PCC

  

  

    
HC during object anticipation: 

    
occipital    lobes and left middle frontal gyrus

  

  

    
HC during food anticipation: 

    
left    IFC and occipital lobes

  

  

    
recAN vs. HC: increased activation due to food stimuli: 

    
precuneus 

  

  

    
recAN vs. HC: decreased activation due to    non-food stimuli:

    
pregenual ACC 

  

  

    
recAN vs. HC: increased activation due to    food anticipation:

    
putamen,    superior gyrus, and medial frontal gyrus

  

  

    
recAN vs. HC: decreased activation due to food anticipation: 

    
IPL 

  

  

    
recAN vs. HC: increased activation due to food stimuli: 

    
IPL, insula, lateral OFC 

  

  

    
recAN vs. HC: decreased activation due to food stimuli: 

    
medial temporal gyrus 

  

  

    
recAN vs. HC: increased activation due to    food anticipation:

    
right ventral anterior insula 

  

  

    
recAN vs. HC: decreased activation due to object anticipation: 

    
right ventral anterior insula 

  

  

    
Correlation between pleasantness    of images rating and brain activation:

  

  

    
HC:    positive correlation with increased activation:

    
insula 

  

  

    
AN: no such relationship 

    
  

  

  

    
Brooks et al. [34] 

    
18 AN (11 rAN, 7 bp AN)/24 HC 

      16-50 yo 

    
Images of food (H and L) and    objects 

    
Instruction to imagine eating/    using what is presented, than rating anxiety. 

      EDE-Q, HADS 

    
rAN:    increased activation due to food stimuli:

    
left    cerebellar vermis and visual cortex, right DLPFC, medial PFC

  

  

    
bpAN:    increased activation due to food stimuli:

    
bilateral    cerebellar vermis, right inferior temporal gyrus

  

  

    
AN vs. HC: increased activation due to    food stimuli:

    
right visual cortex 

  

  

    
AN vs. HC: decreased activation due to    food stimuli:

    
bilateral cerebellar vermis 

  

  

    
rAN vs. HC: increased activation due to    food stimuli:

    
right    visual cortex and DLPFC

  

  

    
rAN vs. HC: decreased activation due to    food stimuli:

    
right    insula, right cerebellar vermis

  

  

    
bpAN vs. HC: increased activation due to food stimuli: 

    
right visual cortex 

  

  

    
bpAN vs. HC: decreased activation due to food stimuli: 

    
left    cerebellar vermis, right insula

  

  

    
rAN vs. bpAN: increased activation due    to food stimuli:

    
bilateral    visual cortex, left ACC and parahippocampal gyrus

  

  

    
rAN vs. bpAN: decreased activation due    to food stimuli:

    
left visual cortex 

  

  

    
Holsen et al. [35] 

    
12AN/10 wrAN/11HC 

    
Images of food (H and L) and    objects before and after meal. 

    
fMRI scanning before and after    meal. Rating appetite and STAI before and after each scanning. EDEQ, BDI,    pleasant-ness rating of pictures. 

    
AN vs. HC: (premeal) decreased    activation due to H stimuli:

    
anterior    insula, amygdala, hypothalamus, hippocampus, OFC

  

  

    
AN vs. HC: (postmeal) decreased    activation due to H stimuli:

    
amygdala, insula 

  

  

    
wrAN vs. HC: (premeal) decreased    activation due to H stimuli:

    
hypothalamus, amygdala, anterior    insula 

  

  

    
AN vs. wrAN: (postmeal) increased    activation due to H stimuli:

    
amygdala 

  

  

    
AN vs. wrAN: (postmeal) decreased    activation due to H stimuli:

    
anterior insula 

  

  

    
Correlation between pleasantness    of H images rating or appetite rating and premeal brain activation: 

  

  

    
HC:    positive correlation between pleasantness of images and brain activation:

    
insula 

  

  

    
HC:    positive relationship between appetite rating and brain activation:

    
insula, amygdala 

  

  

    
wrAN:    positive relationship between appetite rating and brain activation:

    
hypothalamus, amygdala 

  

  

    
Kim et al. [36] 

    
18AN 

      (6rAN, 12 bpAN)/20 BN/20 HC 

    
Images of H food and non-food 

    
Hunger rating before and after    scanning, FCCQ-S, BAI. 

    
AN:    increased activation due to H stimuli:

    
the    left anterior insula, bilateral IFG, right superior frontal gyrus, left ACC,    precuneus and cuneus, bilateral cerebellum

  

  

    
BN:    increased activation due to H stimuli:

    
the    left anterior insula, the left cuneus, bilateral cerebellum

  

  

    
HC:    increased activation due to H stimuli:

    
left    middle frontal gyrus, cuneus and lingual gyrus, right cerebellum

  

  

    
AN vs. HC: increased activation due to    H stimuli:

    
right    IFG, bilateral superior frontal gyrus, left ACC, right cerebellum

  

  

    
AN vs. HC: decreased activation due to    H stimuli:

    
right inferior parietal lobule 

  

  

    
BN vs. HC: increased activation due to    H stimuli:

    
right    middle frontal gyrus, right cerebellum

  

  

    
BN vs. HC: decreased activation due to    H stimuli:

    
right    postcentral gyrus, left inferior parietal lobule

  

  

    
AN vs. BN: increased activation due to    H stimuli:

    
bilateral ACC 

  

  

    
AN vs. BN: decreased activation due to    H stimuli:

    
right middle temporal gyrus 

  

  

    
Functional connectivity between    the left anterior insula and other brain regions:

  

  

    
AN: 

    
right    insula, right IFG, mOFC

  

  

    
Lawson et al. [37] 

    
13AN/9wrAN/13HC 

    
Images of food (H and L) and    objects 

    
Measurement of oxytocin level as    fasted and 3 times after the meal. fMRI before and after the meal. 

      BDI, STAI, EDE-Q 

    
Association between oxytocin level    and brain regions due to food stimuli:

  

  

    
AN vs. HC (premeal): 

    
left    hypothalamus, amygdala and hippocampus, right OFC, bilateral insula

  

  

    
wrAN vs. HC (premeal): 

    
right    insula, bilateral hypothalamus, left amygdala

  

  

    
AN vs. HC (postmeal): 

    
left amygdala and insula 

  

  

    
AN vs. wrAN (postmeal): 

    
right amygdala, bilateral insula 

  

  

    
Brooks et al. [38] 

    
18AN (11 rAN, 7bpAN)/8BN/24HC 

      16-50 yo 

    
Images of food (H and L) and    objects 

    
Instruction to imagine eating/    using what is presented, than rating anxiety. 

      EDE-Q, HADS 

    
AN:    increased activation due to food stimuli:

    
left    visual cortex, cerebellum, right precuneus and DLPFC

  

  

    
rAN:    increased activation due to food stimuli:

    
right    DLPFC and parietal lobe, cerebellum, left visual cortex

  

  

    
bpAN:    increased activation due to food stimuli:

    
bilateral cerebellum, right SMA 

  

  

    
BN:    increased activation due to food stimuli:

    
left    DLPFC, right insula, right visual cortex, left precentral gyrus

  

  

    
HC:    increased activation due to food stimuli:

    
right    insula, right superior and middle temporal gyrus, left caudate, left    cerebellum

  

  

    
AN vs. BN: increased activation due to    food stimuli:

    
parietal lobe and PCC 

  

  

    
AN vs. BN: decreased activation due to    food stimuli:

    
caudate, insula, SMA 

  

  

    
rAN vs. BN: increased activation due to    food stimuli:

    
precentral gyrus 

  

  

    
rAN vs. BN: decreased activation due to    food stimuli:

    
PCC,    ITG, fusiform gyrus, IPL

  

  

    
bpAN vs. BN: increased activation due to    food stimuli:

    
ITG 

  

  

    
bpAN vs. BN: decreased activation due to    food stimuli:

    
PCC, SMA, cerebellum, PHG 

  

  

    
HC vs. BN: increased activation due to    food stimuli:

    
insula, visual cortex 

  

  

    
Cowdrey et al. [39] 

    
15 recAN /16HC 

    
Pictures of moldy strawberry    (aversive) and chocolate (H), matched with taste stimuli 

    
Pleasant-ness rating of pictures.    EDE-Q, BDI, FCPS, SHAPS, STAI, SCID. 

    
recAN vs. HC: increased activation due to    H taste stimuli:

    
ventral striatum, PCC, putamen 

  

  

    
recAN vs. HC: increased activation due to    visual H stimuli:

    
anterior    PFC, occipital cortex, subgenual cingulate/ medial PFC

  

  

    
recAN vs. HC: increased activation due to    H visual and taste stimuli:

    
pallidum 

  

  

    
recAN vs. HC: increased activation due to    aversive taste stimuli:

    
insula, putamen 

  

  

    
recAN vs. HC: increased activation due to    aversive taste and visual stimuli:

    
caudate, DLPFC, ACC, operculum 

  

  

    
Joos et al. [40] 

    
11AN/11HC 

    
Images of food and objects 

    
EDI, BDI, MWT-B, rating of    pictures after scanning. 

    
HC:    activation due to food stimuli:

    
ACC,    bilateral insula, left superior and middle frontal lobe (trends in OFC, MCC,    postcentral gyrus)

  

  

    
AN:    activation due to food stimuli:

    
right    amygdala, precuneus, ACC, MCC, right superior and left middle frontal lobe    (trends to thalamus and lingual gyrus)

  

  

    
AN vs. HC: increased activation due to    food stimuli:

    
right amygdala 

  

  

    
AN vs. HC: decreased activation due to    food stimuli:

    
posterior MCC 

  

  

    
Correlation between disgust    ratings and brain activation:

  

  

    
AN: negative correlation: 

    
right amygdala 

  

  

    
Rothe-mund et al. [41] 

    
12AN/12HC 

    
Images of food (H and L), food    related utensils, neutral objects 

    
SCID, Y-BOCS, TFEQ. 

      VBM and fMRI. 

      Recognition test after scanning,    whether pictures were previously seen or not. 

    
AN:    increased activation due to H stimuli:

    
right    ITG, left middle occipital gyrus, bilateral lingual, inferior occipital gyrus    and precuneus, right cuneus, left culmen, left middle temporal gyrus, right    superior frontal gyrus, left middle frontal gyrus

  

  

    
AN:    increased activation due to L stimuli:

    
right insula 

  

  

    
AN:    decreased activation due to L stimuli:

    
bilateral medial frontal gyrus 

  

  

    
AN:    increased activation due to food related utensils stimuli:

    
right    superior temporal gyrus, left middle frontal gyrus, left claustrum, right    corpus callosum, left supramarginal gyrus, right cingulate gyrus

  

  

    
HC:    increased activation due to H stimuli:

    
right    precuneus and caudate body

  

  

    
HC:    increased activation due to utensils stimuli:

    
right    DLPFC and middle frontal gyrus

  

  

    
AN vs. HC: increased activation due to    H stimuli:

    
right    precuneus and caudate body

  

  

    
Correlations between psychological    tests results and brain regions:

  

  

    
Compulsivity    due to H stimuli correlated with brain activation:

    
the    superior frontal gyrus, inferior frontal gyrus, anterior cingulate cortex,    cingulate gyrus , caudate body, cuneus, pre- and postcentral gyrus

  

  

    
Gizewski et al. [42] 

    
12AN/12 HC 

    
Images of H food and neutral    pictures (IAPS) 

    
Scanning in state of hunger and    satiety. Rating hunger and valence of pictures. SIAB-S, SCID. 

    
AN:    activation in state of hunger:

    
PFC,    central cortices and insula

  

  

    
HC:    activation in state of hunger:

    
ACC, insula 

  

  

    
AN vs. HC: activation in state of    hunger:

    
dPCC 

  

  

    
AN:    activation in state of satiety:

    
left insula 

  

  

    
Association    between food valence judgment and brain activation in AN:

    
insula,    OFC, cingulate cortex and MTG

  

  

    
Santel et al. [43] 

    
13AN/10HC 

      13-21 yo 

    
640 images of H food and objects 

    
Scanning in state of hunger and    satiety. BDI, TFEQ. Rating hunger and valence of pictures. 

    
Food vs.    objects stimuli: 

  

  

    
AN:    activation in state of satiety:

    
right    inferior occipital gyrus and cerebellum (declive), left lingual gyrus and    cerebellum (declive)

  

  

    
AN:    activation in state of hunger:

    
left    cuneus, right fusiform gyrus

  

  

    
AN satiated vs. hungry: 

    
right middle occipital gyrus 

  

  

    
HC:    activation in state of satiety:

    
right    cuneus and middle occipital gyrus, left cuneus and inferior occipital gyrus

  

  

    
HC:    activation in state of hunger:

    
bilateral    lingual gyrus, right fusiform gyrus

  

  

    
HC satiated vs. hungry: 

    
right    ACC, left lateral OFC, left middle temporal gyrus

  

  

    
AN vs. HC: decreased activation in    satiety:

    
left IPL 

  

  

    
AN vs. HC: decreased activation in    hunger:

    
right lingual gyrus 

  

  

    
Association between psychological    tests and brain activation:

  

  

    
“Dietary    restrain” (TFEQ) correlated negatively with:

    
left    IPL, right lingual gyrus

  

  

    
“Disinhibition”    (TFEQ) correlated positively with:

    
left    IPL, right lingual gyrus

  

  

    
BMI correlated positively with: 

    
left IPL 

  

  

    
Uher et al. [44] 

    
16AN (9rAN, 7bpAN)/10BN /19HC 

    
Images of food and objects,    emotional aversive and neutral pictures (IAPS). 

    
MOCI, BDI, rating hunger. 

      Asked to think how presented    pictures make them hungry/ feeling. After scanning rating 

      for pleasant-ness, disgust, fear,    “desire to eat.” 

    
AN:    activation due to food stimuli:

    
left    medial OFC, left ACC, PCC, lateral PFC, right cerebellum

  

  

    
BN:    activation due to food stimuli:

    
left    medial OFC, left ACC, PCC, right cerebellum

  

  

    
HC:    activation due to food stimuli:

    
left    parietal cortex, left lateral PFC, bilateral visual cortex and cerebellum

  

  

    
AN    and BN vs. HC: increased activation    due to food stimuli:

    
left VMPFC 

  

  

    
AN    and BN vs. HC: decreased activation    due to food stimuli:

    
left    lateral PFC, left DLPFC, left IPL, left cerebellum (declive), left occipital    cortex

  

  

    
AN vs. HC: increased activation due to    food stimuli:

    
left    VMPFC, right lingual gyrus

  

  

    
AN vs. HC: decreased activation due to    food stimuli:

    
IPL, cerebellum (declive) 

  

  

    
BN vs. HC: increased activation due to    food stimuli:

    
left    VMPFC, left lingual gyrus, bilateral cerebellum (vermis)

  

  

    
BN vs. HC: decreased activation due to    food stimuli:

    
left    DLPFC, left lateral PFC

  

  

    
AN vs. BN: increased activation due to    food stimuli:

    
right    apical and lateral PFC, right lingual gyrus

  

  

    
AN vs. BN: decreased activation due to    food stimuli:

    
right cerebellum 

  

  

    
rAN vs. HC: increased activation due to    food stimuli:

    
left medial PFC 

  

  

    
bpAN vs. HC: increased activation due to food stimuli: 

    
right    lateral and anterior OFC

  

  

    
rAN vs. bpAN: decreased activation due    to food stimuli:

    
right    anterior PFC and lateral OFC

  










Table 2 off 1:

Stimuli

The main aspect of this review was to analyse cerebral activation due to the presentation of food pictures because such stimuli can be seen as possible symptom provocation. Food pictures categories were either unclassified or divided into high (H) and low-calorie (L). There were also used images of sweet and nonsweet food [21], as well as high and low palatable meals [26]. As this distribution was based on fat and sugar content, it could be compared to high and low-calorie division. When deciding, what kind of object images (as a contrast) should be included, researchers took those with no association to eating. The background of the pictures was as similar as possible (e.g., objects on plates or white circle), so they were matched with food images for arousal and complexity. To enhance comparability between studies, Blechert and colleagues [45] created database of food pictures, with described its features like brightness, contrast within objects, complexity, colours, etc. Images were estimated for number of kilocalories (kcal) and macronutrient composition. Usually, participants were viewing passively presented pictures, but in some studies, to engage them cognitively, they were asked to imagine using/eating items [28,29,34].

Participants´characteristic

While most analyses compared patients with AN to healthy controls (HC), some studies included more categories of subjects, like athletes [31] or participants recovered and weight restored from AN [27-30,32,33,35,37,39]. One study described also acutely ill anorectic patients, but already with normal Body Mass Index (BMI) [26]. Anorectic studies´ participants were usually restricting type, excluding several papers, where some patients were binge-eating/ purging [19,22,31,34,35,38,44] or atypical [21,23].

All included papers concerned female subjects, which could be explained by reports, that only 10-25 % of AN (together with BN) patients are male [46,47] and they are commonly underdiagnosed [48].

In numerous studies [20,22,25,27,30-32,34,35,38,40,43,44] patients were on medications (antidepressants, antianxiety, antipsychotic and antiepileptic medications, amphetamine/ dextroamphetamine). Drug administration often supports psychotherapy of AN [49], mainly due to comorbid depression and anxiety [50]. However, it can also influence functional MRI scans. After exclusion of patients on medications (and those, who at the day of scanning had already gained the weight, so they did not meet all criteria of ED), the remained drug naïve group had increased activation of anterior cingulate cortex (ACC) and medial orbitofrontal cortex (OFC), also decreased activation of inferior parietal lobe (IPL), lateral prefrontal cortex (PFC) and cerebellum. What is more, patients on Selective Serotonin Reuptake Inhibitors (SSRIs) had increased activation of OFC and decreased activation of lateral PFC [44].

Adolescents

We have planned to review functional MRI studies with food stimuli on adolescents, but very few papers considered juvenile in the participants´ group. Only 7 studies focused on minors [17,21,23- 26,43], but two of them referred to functional connectivity [23] and response time to displayed images, not the specific brain region´s activation due to stimuli [17]. Another 2 reports on adults also considered teenagers (from 16 years old) [34,38].

Younger population of anorectic patients were often more occupied with low-calorie food intake than body shape [51], comparing to adult patients. In future, further studies on adolescents with AN are needed and therefore stimuli should be optimized as sensitive for given participants.

Tests

Besides fMRI all studies included also additional psychological and clinical data to their experimental procedures (Table 3). They served mainly as diagnostic tools to define participants, set precise methods or present comorbidity, as anorectic patients often demonstrate dual diagnosis or specific psychological traits [52]. Tests detecting depression or anxiety were explicitly popular. Anxiety is considered both as premorbid trait [53], as well as one of the typical factors of active AN [54,55].

Table 3: Tests and scales used in included studies.





  

    
Diagnostic tests 

  

  

    
SCID-I 

    
Young et al. [20];    Stopyra et al. [22]; Weinbach et al. [17]; Boehm et al. [24]; Kerr    et al. [26]; Scaife et al. [27]; Kullmann et al. [31]; Holsen et al. [30,35];    Lawson et al. [32,37]; Brooks SJ et al. [34,38]; Cowdrey et al. [39];    Rothemund et al. [41]; Gizewski et al. [42] 

  

  

    
SIAB 

    
Boehm et al. [24]; Kerr et    al. [26]; Gizewski et al. [42]

  

  

    
DIPS 

    
Horndasch et al.    (adults) [25] ; Santel et al. (adults) [43]

  

  

    
DISYPS-KJ 

    
Horndasch et al.    (adolescents) [25]; Santel et al. (adolescents) [43]

  

  

    
WAIS/WISC

      WASI-II 

    
Weinbach et al. [17]; Boehm    et al. [24]

  

  

    
MINI-KID 

    
Olivo et al. [23] 

  

  

    
Depression scales 

    
  

  

    
BDI 

    
Horster et al. [19]; Stopyra    et al. [22]; Weinbach et al. [17]; Horndasch et al. [25];    Scaife et al. [27]; Sultson et al. [28]; Holsen et al. [30,35]; Lawson et al.    [32,37]; Oberndorfer et al. [33]; Cowdrey et al. [39]; Santel et al. [43];    Uher et al. [44]

  

  

    
MADRS 

    
Olivo et al. [23] 

  

  

    
HADS 

    
Brooks et al. [34, 38] 

  

  

    
DASS 

    
Young et al. [20] 

  

  

    
PANAS 

    
Young et al. [20] 

  

  

    
PHQ-D 

    
Kullmann et al. [31]

  

  

    
Anxiety and other traits    scales 

  

  

    
STAI 

    
Horster et al. [19]; Ziv et    al. [21]; Weinbach et al. [17]; Scaife et al. [27]; Sultson et al. [28];    Sanders et al. [29]; Kullmann et al. [31]; Oberndorfer et al. [33]; Holsen et    al. [30]; Lawson et al. [37]; Cowdrey et al. [39]

  

  

    
HAM-A 

    
Kerr et al. [26] 

  

  

    
FMPS 

    
Oberndorfer et al. [33]

  

  

    
TCI 

    
Oberndorfer et al. [33]

  

  

    
BIS-11 

    
Oberndorfer et al. [33]

  

  

    
TAS-20 

    
Oberndorfer et al. [33]

  

  

    
Y-BOCS 

    
Rothemund et al. [41]

  

  

    
MOCI 

    
Uher et al. [44]

  

  

    
FCPS 

    
Cowdrey et al. [39]

  

  

    
SHAPS

    
Cowdrey et al. [39]

  

  

    
OCI

    
Weinbach et al. [17] 

  

  

    
Cognitive and behavioural    scales 

  

  

    
BCST 

    
Sultson et al. [28]

  

  

    
BAS/BIS 

    
Kullmann et al. [31]

  

  

    
CES 

    
Kullmann et al. [31]

  

  

    
NART 

    
Scaife et al. [27]

  

  

    
MWT-B 

    
Horster et al. [19]; Joos et    al. [40]

  

  

    
CFT 20 

    
Santel et al. [43]

  

  

    
Food    related and eating disorder specific tests (including eating behaviour tests) 

  

  

    
EDE-Q 

    
Horster et al. [19];    Young et al. [20]; Stopyra et al. [22]; Weinbach et al. [17];    Olivo et al. [23]; Scaife et al. [27]; Sanders et al. [29]; Kullmann et al.    [31]; Brooks et al. [34,38]; Holsen et al. [30, 35]; Lawson et al. [37];    Cowdrey et al. [39] 

  

  

    
EDI-2, EDI-3 

    
Horster et al. [19];    Horndasch et al. [25]; Boehm et al. [24]; Kullmann et al. [31]; Kerr    et al. [26]

  

  

    
EAT-26 

    
Ziv et al. [21] 

  

  

    
TFEQ 

    
Rothemund et al. [41];    Santel et al. [43]

  

  

    
YBC-EDS 

    
Young et al. [20]; Scaife    et al. [27] 

  

  

    
LOFPQ 

    
Scaife et al. [27]

  

  

    
Chocolate 

    
Cowdrey et al. [39]

  

  

    
Hunger rating 

    
Stopyra et al. [22];    Boehm et al. [24]; Holsen et al. [30, 35]; Kullmann et al. [31]; Lawson et    al. [32]; Kim et al. [36]; Gizewski et al. [42]; Santel et al. [43]; Uher et    al. [44]

  

  

    
SCID: Structured Clinical    Interview for DSM Disorders; SIAB: Structured Interview for Anorexic and    Bulimic Disorders; DIPS: Structured Diagnostic Interview for Mental    Disorders; DISYPS-KJ: Diagnostic System for Mental Disorders for Children and    Adolescents; WAIS: Wechsler Adult Intelligence Scale; WISC: Wechsler    Intelligence Scale for Children; WASI-II: Wechsler abbreviated Scale of Intelligence;    MINI-KID: Mini International Neuropsychiatric Interview for Children and    Adolescents; BDI: Beck Depression Inventory; MADRS: Montgomery-�sberg    Depression Rating Scale; DASS: Depression Anxiety Stress Scales; PANAS: Positive    and Negative Affect Schedule; STAI: State-Trait Anxiety Inventory; HAM-A: Hamilton    Anxiety Scale; PHQ-D: Patient Health Questionnaire-Depression Scale; HADS: The    Hospital Anxiety and Depression Scale; FMPS: Frost Multidimensional    Perfectionism Scale; TCI: Temperament and Character Inventory; BIS-11: Barratt    Impulsiveness Scale-11; TAS-20: Toronto Alexithymia Scale-20; Y-BOCS: Yale-Brown    Obsessive Compulsive Scale; MOCI: Maudsley Obsessive-Compulsive Inventory;    FCPS: Fawcett-Clarke Pleasure Scale; SHAPS: Snaith-Hamilton Pleasure Scale;    BCST: Berg Card Sorting Test; Behavioral Activation/Behavioral Inhibition    System scales; CES: Commitment to Exercise Scale; NART: National Adult    Reading Test; MWT-B: Multiple Choice Verbal Comprehension Test; CFT 20: Culture    Fair Intelligence Test; OCI: Obsessive-Compulsive Inventory; EDE-Q: Eating    Disorders Examination-Questionnaire; EAT-26: Eating Attitude Test; TFEQ: Three-Factor    Eating Questionnaire; YBC-EDS: Yale-Brown-Cornell Eating Disorder Scale;    LOFPQ: Leeds-Oxford Food Preference Questionnaire; Chocolate-Rolls-McCabe    Questionnaire for Cravers/Non-Cravers of Chocolate. 

  










Table 3: Tests and scales used in included studies.

Discussion

Adolescents

Research results concerning adolescents are more consistent than those concerning adults, probably due to the larger homogeneity of the group. Viewing food images led to increased activity in the medial prefrontal cortex, the inferior frontal gyrus, the insula, the hippocampus, the fusiform gyrus, the parahippocampal gyrus and the cuneus in anorectic adolescents. The synthetized results of this meta-analysis are presented on the (Figure 2).

Figure 2: Summary of meta-analytic increased activations due to food stimuli in adolescents with AN. Regional labels are only approximate, shown for illustrative purpose. Navy-the medial prefrontal cortex; red-the inferior frontal gyrus; yellow-the insula; white-the hippocampus; green-the fusiform gyrus; orange-the parahippocampal gyrus; blue-the cuneus.

    

    
    

    

    Figure 2: Summary of meta-analytic increased activations due to food stimuli in adolescents with AN. Regional labels are only approximate, shown for illustrative
purpose. Navy-the medial prefrontal cortex; red-the inferior frontal gyrus; yellow-the insula; white-the hippocampus; green-the fusiform gyrus; orange-the
parahippocampal gyrus; blue-the cuneus.

Adults

To summarize, studies concerning anorectic adults revealed enhanced activity due to visual food stimuli in the fusiform gyrus, the inferior frontal gyrus, the lingual gyrus, the medial prefrontal cortex, the right dorsolateral prefrontal cortex, the right angular gyrus. There was deactivation detected in the parahippocampal gyrus, comparing to healthy participants (Figure 3).

Figure 3: Summary of meta-analytic increased activations due to food stimuli in adults with AN. Regional labels are only approximate, shown for illustrative purpose. Blue-the right angular gyrus; navy-the medial prefrontal cortex; red-the inferior frontal gyrus; yellow-the right dorsolateral prefrontal cortex; white-the lingual gyrus; green-the fusiform gyrus. Orange-decreased activity in the parahippocampal gyrus.

    

    
    

    

    Figure 3: Summary of meta-analytic increased activations due to food stimuli in adults with AN. Regional labels are only approximate, shown for illustrative
purpose. Blue-the right angular gyrus; navy-the medial prefrontal cortex; red-the inferior frontal gyrus; yellow-the right dorsolateral prefrontal cortex; white-the
lingual gyrus; green-the fusiform gyrus. Orange-decreased activity in the parahippocampal gyrus.

There were inconsistent reports according influence of visual food stimuli on activation or deactivation of the insula, the amygdala, the hippocampus, the hypothalamus, the anterior cingulate cortex (ACC), the thalamus, the orbitofrontal cortex (OFC), when comparing healthy participants with those with AN. We hypothesize, that contrary results could be caused by heterogeneity of participants in different studies, i.e. according age, duration of illness. Some of these findings are discussed as follows.

Insula

To analyse the results, we focused on the brain areas correlated to different aspects of AN. A primary taste cortex is found in insula, which integrates information about oral stimuli [56,57]. It underlies also interoceptive awareness [57,58] and other food-related processes [57], which are important components of AN psychopathology. Together with amygdala and ACC, insula compounds fear network [59]. Participants with AN were significantly more anxious than HC when watching food pictures [34], what is consistent with insular role in anxiety. The involvement of insula before exposure-based therapy was associated with reduction in food-related anxiety after treatment [20]. In AN insular reaction to high-calorie food images was increased comparing to HC both in adult population [36] and in adolescents [25]. In healthy population, adolescents’ brain activity in the insula (as well as in cingulate and cerebellar regions) was enhanced due to high-calorie food comparing to adult participants [25]. Viewing lowcalorie food pictures may also lead to enhancement of insular activity in AN [41]. It was shown, that even anticipating food pictures causes greater activation in the right ventral anterior insula in recovered AN (recAN), comparing to HC [33]. Although in HC they proved correlation between pleasure caused by tasty food and the insular activity, there is no such correlation in recAN [33].

Varied results were found due to satiation state- in hunger insular activity occurred both in AN and HC [42], or enhanced in HC comparing to AN and recAN [35]. There was found correlation between appetite rating and premeal insular activation in HC [35]. Postmeal, insular reaction to high-calorie stimuli normalised in recAN, but remained enlarged in AN [35,42].

Interestingly, increased activity of the insula was also reported in recAN [29,33] as well as in healthy participants [29,38,40]. This could be explained via its role in taste related reward system [57]. In healthy participants pleasantness of images rating was positively correlated with increased activation of the insula [33,35]. Furthermore, Gizewski and colleagues [42] indicated association between food valence judgment and the insular activation in AN.

Fusiform gyrus

A fMRI study on healthy participants reported, that the response of the fusiform gyrus toward the food images depended on the state of satiety- it was stronger in hunger [60]. As previously shown, difficulties in response inhibition characterising AN patients can be caused by altered ventral attention network [61]. Response inhibition to food stimuli comparing to non-food stimuli enhanced activation of the gyrus fusiform [31]. Increased response for food stimuli in the fusiform gyrus was detected in adult AN [19,22,24]. Interestingly, in state of hunger, the activation in the right fusiform gyrus was enhanced due to food stimuli in both groups of young participants: healthy and anorectic (but only p < .001) [43].

DLPFC

Anorectic adolescents developed higher bilateral activity of dorsolateral prefrontal cortex (DLPFC) and amygdala due to negative stimuli (in general, not food related) [62]. DLPFC is a crucial component of self-control process not only as a whole, but also in food related behaviours. Significant activation of the left DLPFC was detected in group characterised as successful in selfcontrol - those who choose presented healthy but disliked low-calorie food over unhealthy but liked high-calorie food [63]. Other AN specific behaviours, like inhibition to energy intake or motivation on further goals were also associated with DLPFC activity [64]. Reaction of DLPFC in response to the appetitive stimuli remained unclear [34]. Its increased activation could be responsible for cognitive and anxious engagement in food stimuli, as suggested by Brooks and colleagues. Especially, that without cognitive component DLPFC was not activated. Furthermore, DLPFC could inhibit insula and cerebellum, that are normally activated when imaging eating food, which is presented on pictures. On the contrary, Sultson and colleagues described a correlation between anxiety and activity of the left DLPFC during food and non-food processing in HC, but not in AN [28]. Activity of the right DLPFC was negatively correlated with perseverative errors during non-food processing by AN, but positively with non-perseverative errors in HC [28]. On the other hand, right DLPFC demonstrated increased activity in healthy subjects due to high-calorie food and food related utensils images [41]; but also decreased activity in AN comparing to HC due to low-calorie stimuli [27]. When taken together patients with AN and Bulimia Nervosa (BN), they showed decreased activation of left DLPFC due to food stimuli [44].

DLPFC in women is more sensitive to visual hedonic food stimuli [64]. As DLPFC activity was negatively correlated with energy intake, it can provide cognitive control on desire to eat [64]. This conclusion is in line with increased activation of right DLPFC (due to food stimuli) in a restrictive type, but not binge eating AN [34,38]. Patients recovered from AN displayed increased activation of right DLPFC due to aversive taste and visual stimuli [39].

VMPFC

DLPFC influences ventromedial prefrontal cortex (VMPFC) in successful self-control [65]. VMPFC (together with ventral striatum and PCC) is crucial for valuating stimuli [66]. The role of VMPFC in valuating food stimuli was proven by Hare and colleagues, when participants were asked to choose which of viewed food images they would like to eat after scanning [65]. Both people who stayed strict to their diet and those who failed, displayed activation of VMPFC during evaluating food for taste. What is more, VMPFC in participants controlling themselves was also involved in estimation of health impact [65]. Perfectionism and strong self-control are significantly higher in anorectic patients [67]. These findings are in line with increased activation of the left VMPFC in active AN when watching images of food [44].

MCC

On the other hand, the midcingulate cortex (MCC) was positively correlated with failed self-control [63], presenting decreased activation of the posterior MCC due to food stimuli when comparing AN vs. HC [40]. Surprisingly, activation of MCC due to food stimuli was detected in AN, but also as a trend in HC [40].

Amygdala

Part of the fear network is an amygdala [68], region activated in AN when viewing high-calorie food, with increased activation in AN comparing to HC [40]. During adolescence, in response to high palatability stimuli amygdala´s (and subgenual ACC) activation was related to stomach sensation intensity ratings - positively by weight restored AN, but negatively by HC [26]. It was negatively correlated with disgust ratings by anorectic patients [40]. Disfunction of the hypothalamic-pituitary-adrenal (HPA) axis (HPA) and elevation of cortisol and ACTH level occur in AN due to depression, anxiety disorders and long-term starvation, but also independently [69,70]. Cortisol level was associated with amygdala activity changes - with enhanced signal premeal in acute and weight-restored (wrAN) patients, but decreased post meal in AN [32].

In contrary, amygdala activation was decreased due to highcalorie food stimuli in AN vs. HC (pre- and postmeal), wrAN vs. HC (premeal), but still increased in AN comparing to wrAN (postmeal) [35]. In HC and wrAN there was found positive relationship between appetite rating and amygdala activity (premeal) [35]. In fear circuitry amygdala is connected with mPFC [68], which activity was increased due to food visual stimuli in recAN [34,39,44] and significantly correlated with perseverative errors during non-food processing [28].

Hippocampus

The amount of papers describing role of hippocampus in feeding decision is growing recently [71,72]. During food image presentation, AN and recAN showed enhanced activation of hippocampus comparing to HC [29]. Adolescents suffering from AN displayed increased activation of hippocampus due to sweet versus nonfood stimuli [21]. On the contrary, Lawson described hypoactivation of hippocampus in AN vs. HC, but not in recAN vs. HC [37]. Hypothetical reason of changes in hippocampus in AN could be extensive exercising, which is typical behavior for AN. Probably intense physical activity may provide enlargement of the hippocampus, which would be diminished to volume of other anorectic patients after weight restoration [73].

Conclusion

In summary, although there is a growing number of neuroimaging studies concerning pathomechanism of AN, only few of them involved children and adolescents. This noticeable insufficient amount of literature is surprising, considering that AN usually develops in adolescence. There is an urgent need to broaden insight into the neural activity underlying anorexia nervosa in this group of patients. Additionally, it was already pointed by the authors of a replication study, that displayed results were only partially consistent with those from the initial study [19]. A proper number of participants and their homogeneity, along with well-established protocols are features, which are inevitably required to compare any reliably results.

References

  1. American Psychiatric A. Desk Reference to the Diagnostic Criteria from DSM- 5. Washington, DC: American Psychiatric Publishing. 2013.
  2. Jagielska G, I Kacperska. Outcome, comorbidity and prognosis in anorexia nervosa. Psychiatr Pol. 2017; 51: 205-218.
  3. Hilbert A. Childhood Eating and Feeding Disturbances. Nutrients. 2020; 12: 972.
  4. Chang CY, DS Ke, JY Chen. Essential fatty acids and human brain. Acta Neurol Taiwan. 2009; 18: 231-241.
  5. Modan-Moses D, et al. Prospective Longitudinal Assessment of Linear Growth and Adult Height in Female Adolescents with Anorexia Nervosa. J Clin Endocrinol Metab. 2021; 106: e1-e10.
  6. Misra M. Long-term skeletal effects of eating disorders with onset in adolescence. Ann N Y Acad Sci, 2008; 1135: 212-218.
  7. Frank GK. Advances from neuroimaging studies in eating disorders. CNS Spectr. 2015; 20: 391-400.
  8. Arcelus J, et al. Mortality rates in patients with anorexia nervosa and other eating disorders. A meta-analysis of 36 studies. Arch Gen Psychiatry. 2011; 68: 724-731.
  9. Sullivan PF. Mortality in anorexia nervosa. Am J Psychiatry. 1995; 152: 1073- 1074.
  10. Harris EC, B Barraclough. Excess mortality of mental disorder. Br J Psychiatry. 1998; 173: 11-53.
  11. Steinhausen HC. The outcome of anorexia nervosa in the 20th century. Am J Psychiatry. 2002; 159: 1284-1293.
  12. Wenz F, et al. Functional magnetic resonance imaging at 1.5 T: activation pattern in schizophrenic patients receiving neuroleptic medication. Magn Reson Imaging. 1994; 12: 975-982.
  13. Katzman DK, et al. Cerebral gray matter and white matter volume deficits in adolescent girls with anorexia nervosa. J Pediatr. 1996; 129: 794-803.
  14. Gong B, et al. Neuroimaging in Psychiatric Disorders: A Bibliometric Analysis of the 100 Most Highly Cited Articles. J Neuroimaging. 2019; 29: 14-33.
  15. Rees G, et al. Characterizing the relationship between BOLD contrast and regional cerebral blood flow measurements by varying the stimulus presentation rate. Neuroimage. 1997; 6: 270-278.
  16. Zhu Y, et al. Processing of food, body and emotional stimuli in anorexia nervosa: a systematic review and meta-analysis of functional magnetic resonance imaging studies. Eur Eat Disord Rev. 2012; 20: 439-450.
  17. Weinbach N, J Lock, C Bohon. Superior response inhibition to high-calorie foods in adolescents with anorexia nervosa. Behav Res Ther. 2020; 124: 103441.
  18. Moher D, et al. Preferred reporting items for systematic reviews and metaanalyses: the PRISMA statement. Ann Intern Med. 2009; 151: 264-269.
  19. Horster I, et al. A Neglected Topic in Neuroscience: Replicability of fMRI Results with Specific Reference to ANOREXIA NERVOSA. Front Psychiatry. 2020; 11: 777.
  20. Young KS, et al. Exposure to food in anorexia nervosa and brain correlates of food-related anxiety: findings from a pilot study. J Affect Disord. 2020; 274: 1068-1075.
  21. Ziv A, et al. Correlation of Functional Magnetic Resonance Imaging Response to Visual Food Stimuli with Clinical Measures in Adolescents with Restrictive Eating Disorders. J Adolesc Health. 2020; 67: 209-217.
  22. Stopyra MA, et al. The influence of homeostatic mechanisms on neural regulation of food craving in anorexia nervosa. Psychol Med. 2021; 51: 1011- 1019.
  23. Olivo G, et al. Functional connectivity underlying hedonic response to food in female adolescents with atypical AN: the role of somatosensory and salience networks. Transl Psychiatry. 2019; 9: 276.
  24. Boehm I, et al. Subliminal and supraliminal processing of reward-related stimuli in anorexia nervosa. Psychol Med. 2018; 48: 790-800.
  25. Horndasch S, et al. Neural processing of food and emotional stimuli in adolescent and adult anorexia nervosa patients. PLoS One. 2018; 13: e0191059.
  26. Kerr KL, et al. Influence of Visceral Interoceptive Experience on the Brain’s Response to Food Images in Anorexia Nervosa. Psychosom Med. 2017; 79: 777-784.
  27. Scaife JC, et al. Differential activation of the frontal pole to high vs low calorie foods: The neural basis of food preference in Anorexia Nervosa? Psychiatry Res Neuroimaging. 2016; 258: 44-53.
  28. Sultson H, et al. Associations between neural correlates of visual stimulus processing and set-shifting in ill and recovered women with anorexia nervosa. Psychiatry Res Neuroimaging. 2016; 255: 35-42.
  29. Sanders N, et al. Altered food-cue processing in chronically ill and recovered women with anorexia nervosa. Front Behav Neurosci. 2015; 9: 46.
  30. Holsen LM, et al. Abnormal relationships between the neural response to high- and low-calorie foods and endogenous acylated ghrelin in women with active and weight-recovered anorexia nervosa. Psychiatry Res. 2014; 223: 94-103.
  31. Kullmann S, et al. Impaired inhibitory control in anorexia nervosa elicited by physical activity stimuli. Soc Cogn Affect Neurosci. 2014; 9: 917-923.
  32. Lawson EA, et al. Increased hypothalamic-pituitary-adrenal drive is associated with decreased appetite and hypoactivation of food-motivation neurocircuitry in anorexia nervosa. Eur J Endocrinol. 2013; 169: 639-647.
  33. Oberndorfer T, et al. Greater anterior insula activation during anticipation of food images in women recovered from anorexia nervosa versus controls. Psychiatry Res. 2013; 214: 132-141.
  34. Brooks SJ, et al. Thinking about eating food activates visual cortex with reduced bilateral cerebellar activation in females with anorexia nervosa: an fMRI study. PLoS One. 2012; 7: e34000.
  35. Holsen LM, et al. Food motivation circuitry hypoactivation related to hedonic and nonhedonic aspects of hunger and satiety in women with active anorexia nervosa and weight-restored women with anorexia nervosa. J Psychiatry Neurosci. 2012; 37: 322-332.
  36. Kim KR, et al. Functional and effective connectivity of anterior insula in anorexia nervosa and bulimia nervosa. Neurosci Lett. 2012; 521: 152-157.
  37. Lawson EA, et al. Oxytocin secretion is associated with severity of disordered eating psychopathology and insular cortex hypoactivation in anorexia nervosa. J Clin Endocrinol Metab. 2012; 97: E1898-E1908.
  38. Brooks SJ, et al. Differential neural responses to food images in women with bulimia versus anorexia nervosa. PLoS One. 2011; 6: e22259.
  39. Cowdrey FA, et al. Increased neural processing of rewarding and aversive food stimuli in recovered anorexia nervosa. Biol Psychiatry. 2011; 70: 736- 743.
  40. Joos AA, et al. Amygdala hyperreactivity in restrictive anorexia nervosa. Psychiatry Res. 2011; 191: 189-195.
  41. Rothemund Y, et al. Compulsivity predicts fronto striatal activation in severely anorectic individuals. Neuroscience. 2011; 197: 242-250.
  42. Gizewski ER, et al. Influence of satiety and subjective valence rating on cerebral activation patterns in response to visual stimulation with high-calorie stimuli among restrictive anorectic and control women. Neuropsychobiology. 2010; 62: 182-192.
  43. Santel S, et al. Hunger and satiety in anorexia nervosa: fMRI during cognitive processing of food pictures. Brain Res. 2006; 1114: 138-148.
  44. Uher R, et al. Medial prefrontal cortex activity associated with symptom provocation in eating disorders. Am J Psychiatry. 2004; 161: 1238-1246.
  45. Blechert J, et al. Food-pics: an image database for experimental research on eating and appetite. Front Psychol. 2014; 5: 617.
  46. Weltzin TE, et al. Eating disorders in men: Update. Journal of Men’s Health & Gender. 2005; 2: 186-193.
  47. Hudson JI, et al. The prevalence and correlates of eating disorders in the National Comorbidity Survey Replication. Biol Psychiatry. 2007; 61: 348-358.
  48. Strother E, et al. Eating disorders in men: underdiagnosed, undertreated, and misunderstood. Eat Disord, 2012; 20: 346-355.
  49. Frank GK, ME Shott. The Role of Psychotropic Medications in the Management of Anorexia Nervosa: Rationale, Evidence and Future Prospects. CNS Drugs, 2016; 30: 419-442.
  50. https://aps.onlinelibrary.wiley.com/doi/abs/10.1111/j.1742-9552.2011.00034.x
  51. Herpertz-Dahlmann B, B Dahmen. Children in Need-Diagnostics, Epidemiology, Treatment and Outcome of Early Onset Anorexia Nervosa. Nutrients. 2019; 11: 1932.
  52. Marucci S, et al. Anorexia Nervosa and Comorbid Psychopathology. Endocr Metab Immune Disord Drug Targets. 2018; 18: 316-324.
  53. Kaye WH, et al. Nothing tastes as good as skinny feels: the neurobiology of anorexia nervosa. Trends Neurosci. 2013; 36: 110-120.
  54. Lulé D, et al. Anorexia nervosa and its relation to depression, anxiety, alexithymia and emotional processing deficits. Eat Weight Disord. 2014; 19: 209-216.
  55. Schulze UM, et al. Trait anxiety in children and adolescents with anorexia nervosa. Eat Weight Disord. 2009; 14: e163-e168.
  56. Roberts CA, et al. A Systematic Review and Activation Likelihood Estimation Meta-Analysis of fMRI Studies on Sweet Taste in Humans. J Nutr. 2020; 150: 1619-1630.
  57. Frank S, et al. Food related processes in the insular cortex. Front Hum Neurosci. 2013; 7: 499.
  58. Craig AD. How do you feel--now? The anterior insula and human awareness. Nat Rev Neurosci. 2009; 10: 59-70.
  59. K Holzschneider, C Mulert. Neuroimaging in anxiety disorders. Dialogues Clin Neurosci. 2011; 13: 453-461.
  60. LaBar KS, et al. Hunger selectively modulates corticolimbic activation to food stimuli in humans. Behav Neurosci. 2001; 115: 493-500.
  61. Collantoni E, et al. Functional connectivity correlates of response inhibition impairment in anorexia nervosa. Psychiatry Res Neuroimaging. 2016; 247: 9-16.
  62. Seidel M, et al. Processing and regulation of negative emotions in anorexia nervosa: An fMRI study. Neuroimage Clin. 2018; 18: 1-8.
  63. Chen F, et al. Increased BOLD Signals in dlPFC is Associated with Stronger Self-Control in Food-Related Decision-Making. Front Psychiatry. 2018; 9: 689.
  64. Cornier MA, et al. Sex-based differences in the behavioral and neuronal responses to food. Physiol Behav. 2010; 99: 538-543.
  65. Hare TA, et al. Self-control in decision-making involves modulation of the vmPFC valuation system. Science. 2009; 324: 646-648.
  66. Clithero JA, Rangel A. Informatic parcellation of the network involved in the computation of subjective value. Soc Cogn Affect Neurosci. 2014; 9: 1289- 1302.
  67. Dahlenburg SC, et al. Anorexia nervosa and perfectionism: A meta-analysis. Int J Eat Disord. 2019; 52: 219-229.
  68. Marek R, et al. The amygdala and medial prefrontal cortex: partners in the fear circuit. J Physiol. 2013; 591: 2381-2391.
  69. Lawson EA, et al. Adrenal glucocorticoid and androgen precursor dissociation in anorexia nervosa. J Clin Endocrinol Metab. 2009; 94: 1367-1371.
  70. Schorr M, KK Miller. The endocrine manifestations of anorexia nervosa: mechanisms and management. Nat Rev Endocrinol. 2017; 13: 174-186.
  71. Kanoski SE, HJ Grill. Hippocampus Contributions to Food Intake Control: Mnemonic, Neuroanatomical, and Endocrine Mechanisms. Biol Psychiatry. 2017; 81: 748-756.
  72. Stevenson RJ, et al. Hippocampal-dependent appetitive control is impaired by experimental exposure to a Western-style diet. R Soc Open Sci. 2020; 7: e191338.
  73. Beadle JN, et al. Larger hippocampus size in women with anorexia nervosa who exercise excessively than healthy women. Psychiatry Res. 2015; 232: 193-199.

Citation: Dabkowska-Mika A, Steiger R, Gander M, Sevecke K and Gizewski ER. Systematic Review of fMRI Studies with Visual Food Stimuli in Anorexia Nervosa. Austin J Nutr Metab. 2021; 8(4): 1115.