Cingulate volume abnormalities in emerging psychosis.

Abstract

Neuroanatomical abnormalities, including cingulate cortex volume abnormalities, are a common feature in psychosis. However, the extent to which these are related to a vulnerability to psychosis, as opposed to the disorder per se, is less certain. AIM UND HYPOTHESES: The aim of the present study is to compare cingulate gray matter volumes in different stages of psychosis. We reviewed previous studies of subjects in a prodromal stage of psychosis and tested cingulate volume changes during the transition to psychosis.
A cross-sectional MRI study of manually traced cingulate gray-matter volumes in 37 individuals with an at risk mental state (ARMS) for psychosis, 23 individuals with a first-episode psychosis (FEP), and 22 healthy controls (HC) was performed using a 1.5 T MRI-scanner. 16 of 37 ARMS individuals (43 %) developed psychosis during follow up (ARMS-T), whereas 21 did not (ARMS-NT). The mean duration of follow up in ARMS was 25.1 months. 8 cingulate subregions were analysed in a region-of-interest analysis.
Compared to HC, subjects with an ARMS had significantly reduced left caudal anterior cingulate cortex volume (p < 0.027). This finding was also evident at a trend level (p: 0.069) in FEP patients. Within ARMS, the ARMS-T group showed a significantly reduced whole right cingulate cortex (p: 0.036), right subgenual cingulate cortex (p: 0.036) and right posterior cingulate cortex (p: 0.012) compared to ARMS-NT.
These results suggest that the at risk mental state is associated with cingulate volume reductions, in particular in the left caudal anterior cingulate cortex (CACC). These abnormalities do not only seem to occur with transition to psychosis, but may be a correlate of an increased vulnerability to psychosis.

Full text

Page 1

CINGULATE VOLUME ABNORMALITIES IN EMERGING PSYCHOSIS
(RUNNING TITLE: CINGULATE ABNORMALITIES IN PSYCHOSIS)
Michel Röthlisberger1,2, Anita Riecher-Rössler1, Jacqueline Aston1, Paolo Fusar-Poli3, Ernst-Wilhelm
Radü2, *Stefan Borgwardt1,2,3
1) Department of Psychiatry, c/o Universitätsspital Basel, University of Basel, Petersgraben 4, CH-4031
Basel
2) Medical Image Analysis Centre, University of Basel, Schanzenstrasse 55, CH-4031 Basel,
Switzerland
3) Institute of Psychiatry, King’s College London, De Crespigny Park, SE5 8AF London, UK
Corresponding author:
Dr Stefan Borgwardt
Department of Psychiatry
c/o University Hospital Basel
Petersgraben 4
CH-4031 Basel
Switzerland
Phone: 0041613286126, Fax: 0041612654588
e-mail: sborgwardt@uhbs.ch

Page 2

ABSTRACT (250 WORDS)
Background
Neuroanatomical abnormalities, including cingulate cortex volume abnormalities, are a common feature
in psychosis. However, the extent to which these are related to a vulnerability to psychosis, as opposed
to the disorder per se, is less certain.
Aim und Hypotheses
The aim of the present study is to compare cingulate gray matter volumes in different stages of
psychosis. We reviewed previous studies of subjects in a prodromal stage of psychosis and tested of
cingulate volume changes during the transition to psychosis.
Methods
A cross-sectional MRI study of manually traced cingulate gray-matter volumes in 37 individuals with an
At Risk Mental State (ARMS) for psychosis, 23 individuals with First-Episode Psychosis (FEP), and 22
Healthy Controls (HC) was performed using a 1.5T MRI-Scanner. 16 of 37 ARMS individuals (43 %)
developed psychosis during follow up (ARMS-T), whereas 21 did not (ARMS-NT). The mean duration
of follow up in ARMS was 25.1 months. 8 cingulate subregions were analysed in a region-of-interest
analysis.
Results
Compared to HC, subjects with an ARMS had significantly reduced left caudal anterior cingulate cortex
volume (p<0.027). This finding was also evident at a trend level (p: 0.069) in FEP patients. Within the
ARMS, ARMS-T group showed significantly reduced whole right cingulate cortex (p: 0.036), right
subgenual cingulate cortex (p: 0.036) and right posterior cingulate cortex (p: 0.012) compared to
ARMS-NT.
2

Page 3

Discussion
These results suggest that the at-risk mental state is associated with cingulate volume reductions in
particular in the left caudal anterior cingulate cortex (CACC). These abnormalities do not only seem to
occur with transition to psychosis, but may be a correlate of an increased vulnerability to psychosis.
3

Page 4

1. INTRODUCTION
Talairach (1993) describes the cingulate gyrus as follows: Bounded superiorly and in the front of the
corpus callosum by the callosomarginal sulcus, the cingulate gyrus extends under the rostrum of the
corpus callosum into the subcallosal gyrus, and behind the splenium into the parahippocampal gyrus
through a narrow passage, the isthmus (Figure 1). It forms the upper part of the great limbic lobe of
Broca [1].
Please insert figure 1: Cingulate Cortex.bmp with legend [53]
The cingulate gyrus is composed of different cytoarchitectural areas, which connect the limbic system
with the neocortex. The several subdivisions have many functions such as emotional, cognitive,
attentional, nociceptive and motoric processing. The anterior cingulate cortex (ACC) is an integral
component of those subdivisions, which controls affective and cognitive functions. The central task of
this area is the modulation of internal emotional responses. The ACC has anatomical connections to the
dorsolateral prefrontal cortex (DLPFC), motoric areas and the thalamus, depending on the specific
subdivision. Those cytoarchitectural subdivisions are the caudal anterior cingulate cortex (CACC,
cognitive subdivision) with strong reciprocal connections to prefrontal, parietal, premotor and
supplementary areas on one side, and the rostral anterior cingulate cortex (RACC, affective subdivision)
with connections to nucleus accumbens, amygdala, insular cortex, hippocampus and orbitofrontal cortex
on the other hand [2, 3].
The ACC is therefore an integral component for social recognition, mentalizing and visualizing. This
area is mainly activated by emotional stimuli [3, 4]. The “cognitive subdivision” is represented by the
CACC, which is involved in the initiation of action, selective perception [5], selection and monitoring
of conflicting responses and error detection [6]. The latter has been proven by studies using the stroop
task test [7]. The RACC (affective subdivision) seems to be increasingly involved in the modulation of
4

Page 5

emotional reactions, which evaluates the salience of motivational and emotional information [3, 8]. The
subgenual cingulate cortex (SCC), also called Brodmann’s area 25, has strong connections to structures
that control emotional behaviour, mood and autonomic reactions to stressors [9]. The posterior cingulate
cortex (PCC) is activated by emotional and non-emotional stimuli and plays an important role in
memory access, visual and spatial orientation [2, 10, 11].
Findings from post mortem neuropathology [12, 13] and structural magnetic resonance imaging of
subjects at genetic [14] or clinical high risk of psychosis [15-19] show that ACC is involved in the
pathogenesis and the etiology of schizophrenia (for review[20-25]: PET studies [26] have brought
increased glutamatergic metabolites [13] and deficits in membrane phospholipids [27] in this region into
focus. PET- and SPECT studies with non-treated schizophrenia patients also showed a negative
correlation between symptoms such as suspiciousness, hallucinations and delusions with regional blood
flow (RBF) in the ACC [28]. Furthermore, Choi et al. (2005) showed, that a smaller volume of the
caudal anterior cingulate gyrus significantly correlated with more severe positive symptoms of
schizophrenia [29]. A number of other studies reported functional deficits of the ACC in FEP individuals
such as reduced activation during verbal fluency [30, 31], executive control task [32], manipulation
phase of working memory [33]. Qualitatively similar abnormal deficits were found in ARMS subjects
[15, 16, 26, 34]. Also a significant activation of the ACC during acoustic verbal hallucination has been
shown [35].
Individuals with ARMS have an increased vulnerability for psychosis. Around 35 % of such subjects
develop psychosis within 12 months, although the proportion has varied between studies [36-38]. It is
well accepted that a dysfunction of cingulate brain region may be a core mechanism in early stage
psychosis involving a failure to monitor internally generated actions [33, 39]. The anterior cingulate
cortex (ACC) volume was investigated in three studies Melbourne group [40-42]. Yücel [40] found no
differences in any of the ACC surface morphological measures between HR-T and HR-NT. Another
study showed a trend towards left hemispheric reduced paracingulate sulcus folding and frequent
cingulate sulcus interruptions in HR subjects, with no differences between HR-T and HR-NT subjects,
5

Page 6

in line with the above findings [41]. Fornito [42] used a surface-based anterior cingulate parcellation
technique and reported that regional thinning of the ACC is a significant predictor of the time to
psychosis onset. They found a bilateral thinning of the rostral paralimbic ACC in HR-T compared to
HC. In a voxel-based morphometry study, Borgwardt et al. (2007) confirmed the hypothesis of reduced
gray matter volumes in a cluster in a midline region that includes the posterior cingulate cortex and the
precuneus in ARMS individuals compared to healthy controls [43].
The aim of this study is to use MRI data of individuals with an ARMS to clarify structural abnormalities
of the Cingulum, which is presumed to be involved in early stages of this disease. Baseline gray matter
volumes of the Cingulum in different states of schizophrenia are compared, particularly ARMS
individuals with transition to psychosis (HR-T), ARMS individuals without transition to psychosis (HR-
NT), First-episode Psychosis (FEP) and healthy controls (HC). In particular we tested the hypothesis
that cingulate abnormalities are associated with an ARMS. Secondly, we tested whether these
volumetric abnormalities were predictive for a future transition to psychosis. By segmenting the
cingulate cortex into subdivisions, particularly subgenual cingulate cortex (SCC), rostral and caudal
anterior cingulate cortex (RACC and CACC) and posterior cingulate cortex (PCC) we expected to
provide evidence for a stage-related distribution of the volume-changes to substantiate a theory of
change progression from rostral to caudal.
6

Page 7

Table 1: Cingulate MRI findings in the At Risk Mental State (ARMS)[15, 29, 42-45]
Baseline MRI Findings
N MRI-Method Schizophrenic vs. HCARMS-T vs.
ARMS-NT
[no control group]Choi et al. 200422 Schizophrenic
Patients vs. 22
Healthy Controls
ROI analysis with
baseline MRI for
rostral anterior
cingulate gyrus,
the caudal anterior
cingulate gyrus,
the orbitofrontal
cortex, the caudate
and the thalamus
ROI analysis with
baseline MRI for
anterior cingulate
cortex (ACC)
volumes
Volumetric reduction
of the right caudal
anterior cingulate
gyrus was observed in
patients with schizo-
phrenia as compared
with the normal
controls
Fujiwara et al.
2007
20 Schizophrenic
Patients vs. 20
Healthy Controls
reduced ACC volume,
decreased fractional
anisotropy in the
anterior cingulum
bilaterally and a poorly
developed
paracingulate/cingulat
e sulcus in the left
hemisphere
[no control group]
NMRI-MethodARMS vs. HCARMS-T vs.
ARMS-NT
Converters had
smaller gray
matter volume in
the right medial
temporal, lateral
temporal, inferior
frontal cortex, and
in the cingulate
bilaterally
less gray matter
volume in the
right insula,
inferior
frontal and
superior temporal
gyrus
RACC, rostral
paralimbic, limbic
and paralimbic
subcallosal
regions show
significant
thinning in
ARMS-T vs.
ARMS-NT
Pantelis et al.
2003
75 ARMS (23
ARMS-T vs. 52
ARMS-NT)
Voxel-based
morphometry
(VBM) MRI-
analysis
[no control group]
Borgwardt et
al. 2007
35 ARMS, 25
FEP, and 22 HC
Voxel-based
morphometric
(VBM) study
using 1.5-T-MRI
Converters had smaller
smaller left insula,
superior temporal
gyrus, cingulate gyrus
and precuneus
Fornito et
al.2008
35 ARMS-T vs.
35 ARMS-NT;
33 HC
ROI analysis with
baseline MRI for
ACC
morphometry
ARMS-T bilateral
thinning of a rostral
paralimbic ACC
region, ARMS-NT
individuals had a
relative thickening of
dorsal and rostral
limbic areas
MRI-MethodFEP vs. HCARMS-T vs.
ARMS-NT
7

Page 8

Koo et al.200839 FEP vs. 40
HC
FEP showed
significantly smaller
left subgenual, left and
right affective, right
cognitive, and right
posterior cingulate
gyrus gray matter
subregions. Less
asymmetric
paracingulate pattern.
[no control group]
ARMS: At-Risk-Mental-State, ARMS-T: At-Risk-Mental-State with Transition to psychosis; ARMS-
NT: At-Risk-Mental-State without Transition to psychosis
FEP: First Episode Psychosis; HC: healthy controls; ROI: Region of interest
8

Page 9

2. METHODS
2.1 Participants
The MRI data were collected in the context of the Basel Early-detection-of-Psychosis (Früherkennung
von Psychosen: FePsy) and as part of naturalistic, prospective research program (Prediction and early
detection of schizophrenia - a prospective multilevel approach), supported by the Swiss National
Science Foundation (No. 3200-057216-99; 3200-057216/3). The Basel ethics committee approved all
aspects of the study and written informed consent was obtained from each participant.
Subjects with an ARMS and patients experiencing their FE of psychosis were recruited from a service
area covering 200.000 inhabitants in and around Basel, Switzerland, through a specialized clinic at the
Psychiatric Outpatient Department, University Hospital in Basel. The screening has been described in
detail elsewhere [46, 47]. Briefly, the Basel Screening Instrument for Psychosis (BSIP) was used, a 46-
item checklist based on risk factors or early signs of psychosis, i.e. “prodromal” symptoms, social
decline, previous psychiatric disorders, drug abuse or genetic risk for psychosis [46] and also assesses
severity of prepsychotic symptoms. It allows the identification of ARMS individuals similar to the
PACE criteria [48]. The BSIP was constructed as a screening to identify those at risk and is followed by
a more extensive early detection interview in a next step [46] . The family history of psychosis was
obtained using a semi-structured interview from the subject and, whenever possible, a first-degree
relative. The frequency of current and previous alcohol use was estimated using a semi-structured
interview. To assess the premorbid IQ we used the MWT, an established measure in German-speaking
subjects [49]. All assessments were conducted by experienced psychiatrists who underwent regular
training.
2.1.1 Inclusion criteria
9

Page 10

2.1.1.1 At Risk Mental State (ARMS) Group
The ARMS group n=37 in this study fulfilled PACE criteria, similar to previous MRI studies on this at
Risk sample [19, 43, 50-55]. Inclusion thus required one or more of the following: a) “attenuated”
psychotic symptoms, b) brief limited intermittent psychotic symptoms (BLIPS), or c) a first degree
relative with a psychotic disorder plus at least two indicators of a clinical change, such as a marked
decline in social or occupational functioning. Inclusion because of “attenuated” psychotic symptoms
required scores of 2 or 3 on the hallucination item, 3 or 4 on the unusual thought content or
suspiciousness items of the BPRS for at least several times a week and persisting for more than one
week. Inclusion because of BLIPS required scores of 4 or above on the hallucination item or 5 or above
on the unusual thought content, suspiciousness or conceptual disorganization items of the BPRS, with
each symptom lasting less than one week before resolving spontaneously. 16 out of 37 ARMS have so
far made the transition to schizophrenia (HR-T: n=16 / HR-NT: n=19).
2.1.1.2 First Episode Psychosis (FEP) Group
The FE group (n = 23) was defined as subjects who met the operational criteria for first episode
psychosis described by Yung et al. (1998) [48].
2.1.1.3 Healthy Volunteers as Control Group
Healthy volunteers (n = 22) were recruited from the same geographical area as the other groups through
local advertisements [37, 56, 57]. These individuals had no current psychiatric disorder, no history of
psychiatric illness, head trauma, neurological illness, serious medical or surgical illness, substance
abuse, and no family history of any psychiatric disorder as assessed by an experienced psychiatrist in a
detailed clinical interview. After complete description of the study to the subjects, written informed
consent was obtained.
10

Page 11

2.1.2 Exclusion criteria
The following exclusion criteria applied to all groups: history of previous psychotic disorder (treated
with major tranquilizers for more than 3 weeks); psychotic symptomatology clearly due to ‘organic’
brain disease or substance abuse according to ICD-10 research criteria; psychotic symptoms clearly
associated with an affective psychosis or a borderline personality disorder; age under 18 years;
inadequate knowledge of the German language; and IQ less than 70.
After these exclusion criteria had been applied, subjects were assessed using the ‘Basel Screening
Instrument for Psychosis’ (BSIP) [46], the Brief Psychiatric Rating Scale (BPRS) [58, 59], and the Scale
for the Assessment of Negative Symptoms (SANS) [60]. The BSIP was used to evaluate ‘prodromal’
symptoms (defined according to DSM-III-R) occurring in the last 5 years; nonspecific ‘prodromal’ signs
[61] in the last 2 years; previous or current (pre-)psychotic symptoms, psychosocial functioning over the
last 5 years, substance dependency; and psychotic disorders among first and second degree relatives
[46] and to apply the operational criteria of an ARMS resp. FE according to Yung et al. ‘98.
2.1.3 Clinical Follow-up
The ARMS subjects were followed up at monthly intervals during the first year, at 3-month intervals
during the second and third year and annually thereafter. At each assessment, subjects were examined
using the BPRS. The criteria for transition to psychosis were those defined by Yung et al. (1998) [48].
2.2 Structural MRI
2.2.1 MRI Image Acquisition
Subjects were scanned using a Siemens (Erlangen, Germany) Magnetom Vision 1.5 T scanner at the
University Hospital Basel. Head movement was minimized by foam padding and velcro straps across
the forehead and chin. A three-dimensional volumetric spoiled gradient recalled echo sequence
generated 176 contiguous, 1 mm thick sagittal slices. Imaging parameters were: time-to-echo, 4 msec;
11

Page 12

time-to-repetition, 9.7 msec; flip angle, 12; matrix size, 200 x 256; field of view, 25.6 x 25.6 cm matrix;
voxel dimensions, 1.28 x 1 x 1 mm.
2.2.2 Cingulate volume measurement
We focused upon anterior (ACC) and posterior cingulate cortex (PCC) both right and left as our Region-
of-interest (ROI). ACC underwent a further segmentation in functional subdivisions, particularly rostral
anterior (RACC), caudal anterior (CACC) and subgenual cingulate cortex (SCC), both right and left.
Cingulate volume was calculated by summing up all the marked voxels, using the software for medical
imaging amira™, which displays all three planes simultaneously. To reach an equal tracing of those
ROI’s, for each and every patient the same approach was used including clearly defined landmarks and
lines. Tracing was made individually with a mouse-driven cursor for each sagittal and coronal plane,
only one person was involved to avoid differences between individual samples (MR).
Preceding anatomical definition of the ACC and PCC was made by using Talairach Co-Planar
Stereotaxic Atlas [1]. Before tracing began, several reference lines needed to be drawn on the most
medial sagittal slice of each hemisphere. Two landmarks where used to reach a standard segmentation
Figure 2):
•
CA-CP-Line: this line passes through the superior edge of the anterior commissure and the
inferior edge of the posterior commissure. It follows a path essentially parallel to the
hypothalamic sulcus, dividing the thalamic from the sub-thalamic region. This line defines the
horizontal plane [1].
•
VCA-Line: this line is a vertical transversing the posterior margin of the anterior
commissure.
This line is the basis for the vertical plane [1].
Please insert figure 2: Reference Lines.bmp with legend [53]
12

Page 13

By setting the VCA-line with the grid-function offered by the medical imaging software amira™ at the
most superior-posterior edge of the Commissura anterior, we were able to reach a reference-associated
segmentation of the cingulate gyrus in an anterior and a posterior part (Figure 3).
Please insert figure 3: Cingulate Subdivisions (legend).bmp
For a standard definition of the subdivision of the ACC, a line, defined by the coronal plane passing
through the most anterior tip of the inner surface of the genu of the corpus callosum, was used as the
border between the RACC and the CACC on one hand. A different line, which was set by using the
coronal plane passing through the most anterior tip of the outer surface of the genu of the corpus
callosum, served as border between the RACC and the SCC (Figure 4 and 5). This was described before
by Crespo-Facorro et al. (1999) [62].
Please insert figure figure 4: Dividing right Cingulum (legend).bmp and figure 5: Dividing left
Cingulum (legend).bmp
The PCC was bounded by the VCA-line on the one side (as mentioned to be the border between ACC
and PCC), and by a line, which was found by the coronal plane touching the most posterior tip of the
outer surface of the splenium of the corpus callosum. This line was used to have an equal reference for
the dorsal border of the PCC.
Using the most sagittal plane of each hemisphere and considering the reference lines described before,
the four RIO’s (both left and right) were marked by with a mouse-driven cursor in eight different
colours (Figure 4). The paracingulate gyrus was considered to be a part of the anterior cingulate gyrus,
as assumed commonly and practiced by several studies before [45, 62]. For an accurate assessment and
tracing of the gray matter associated to the overall eight different ROI’s, tracing was made in a serial of
coronal planes starting on the first plane showing marked RACC tissue (defined in the sagittal plane
before) and was continued caudally. With this strategy, the area defined in the sagittal plane was
13

Page 14

consequently extended in each coronal plane to gain a three-dimensional information of the cingulate
gray matter volume. The deepest point of the callosal sulcus and the most medial point of the dorsal
bank of the cingulate sulcus were used as the inner and the outer boundaries of the ACC and the PCC in
each coronal slice. If there was a paracingulate gyrus, the next superior sulcus associated to its gray
matter was used as upper border (Figure 6).
Please insert figure 6: Coronal tracing (without legend).bmp
2.2.3 Intra-Rater Reliability
To assess the intra-rater reliability, a manual segmentation in ten consecutive cases within two weeks
was accomplished by the same person (MR). The intra-rater reliability ranged from κleft = 0.93 to κright =
0.95, mean value of κtotal = 0.94. An inter-rater reliability could not be calculated as there was only one
tracer (MR). The researcher who traced the cingulate volumes was blind to the group status at any time
of the study. Once the gray matter of the cingulate cortex had been traced, volumes (in mm3) were
calculated by computing the number of voxels from each traced image.
2.3 Statistical analysis
Statistical analyses were performed with the Statistical Package for the Social Sciences, version 16
(SPSS© 16.0 for Windows, Rel, 16.0, SPSS© Inc., Chicago, Illinois). Clinical and ROI data were
compared using the one-way analysis of variance (ANCOVA) adjusted for age, sex, and gray matter
volume, Student’s t-tests for psychopathology scores (interval-scaled or continuous variables), chi-
square test and Fisher’s exact test for categorical variables. The following ROS’s were analysed: tCC:
total Cingulate Cortex; tCCr : total Cingulate Cortex right; tCCl: total Cingulate Cortex left; tACCr:
total Anterior Cingulate Cortex right; tACCl: total Anterior Cingulate Cortex left; CACCr: Caudaler
Anterior Cingulate Cortex right; CACCl: Caudal Anterior Cingulate Cortex left; RACCr: Rostral
Anterior Cingulate Cortex right; RACCl: Rostral Anterior Cingulate Cortex left; SCCr: Subgenual
14

Page 15

Cingulate Cortex right; SCCl: Subgenualer Cingulate Cortex left; PCCr: Posterior Cingulate Cortex
right; PCCl: Posterior Cingulate Cortex left. All pairwise comparisons used Bonferroni adjustment for
multiple testing was applied. To test our hypothesis, two sets of analysis were performed: Analysis 1:
comparison of ARMS, FEP and HC; Analysis 2: ARMS-T, ARMS-NT, FEP and HC. Within each
analysis, patient groups were compared with the healthy control group. Results of statistical tests are
given as mean ± standard deviation (SD). The level of statistical significance was set to p < 0.05 a trend
was considered p < 0.1.
15