Accepted Manuscript Title: Abnormal Functional Connectivity Density in Psychogenic Non-epileptic Seizures Author: Ju-Rong Ding Dongmei An Wei Liao Guo-Rong Wu Qiang Xu Dong Zhou Huafu Chen PII: DOI: Reference:
S0920-1211(14)00137-5 http://dx.doi.org/doi:10.1016/j.eplepsyres.2014.05.006 EPIRES 5156
To appear in:
Epilepsy Research
Received date: Revised date: Accepted date:
14-11-2013 14-4-2014 6-5-2014
Please cite this article as: Ding, J.-R., An, D., Liao, W., Wu, G.R., Xu, Q., Zhou, D., Chen, H.,Abnormal Functional Connectivity Density in Psychogenic Non-epileptic Seizures, Epilepsy Research (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.05.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Title page Title: Abnormal Functional Connectivity Density in Psychogenic Non‐epileptic
ip t
Seizures
cr
Running title: Abnormal FCD in PNES
us
Authors: Ju‐Rong Ding1,2, Dongmei An3, Wei Liao4,5, Guo‐Rong Wu1, Qiang Xu5, Dong
an
Zhou3*, Huafu Chen1*
M
1
Key Laboratory for Neuroinformation of Ministry of Education, School of Life Science
2
te
610054, PR China.
d
and Technology, University of Electronic Science and Technology of China, Chengdu
Ac ce p
Institute of Automation and Electronic Information, Sichuan University of Science
and Engineering, Zigong 643000, PR China.
3
Department of Neurology, West China Hospital of Sichuan University, Chengdu
610041, PR China.
4
Center for Cognition and Brain Disorders and the Affiliated Hospital, Hangzhou
Normal University, Hangzhou 310015, PR China. 5
Department of Medical Imaging, Nanjing Jinling Hospital, Clinical School, Medical
College, Nanjing University, Nanjing 210002, PR China. 1
Page 1 of 37
*Corresponding authors: Huafu Chen, School of Life Science and Technology, University of Electronic Science
ip t
and Technology of China, Chengdu 610054, PR China. Fax: 86‐28‐83208238. E‐mail: [email protected].
cr
Dong Zhou, Department of Neurology, West China Hospital of Sichuan University,
us
Chengdu 610041, PR China. Fax: 86‐28‐85422549. E‐mail: [email protected].
Ac ce p
te
d
M
an
2
Page 2 of 37
Abstract Purpose: Psychogenic non‐epileptic seizures (PNES) are paroxysmal behaviors that
ip t
resemble epileptic seizures but lack abnormal electrical activity. Some neuroimaging studies have reported that PNES exhibits aberrant functional connectivity in specific
cr
brain networks. Thus, advanced neuroimaging technologies may aid clinical diagnosis
us
and treatment of PNES.
Methods: We investigated changes in brain functional connectivity in 18 patients
an
with PNES and 20 healthy controls. Functional connectivity density mapping (FCDM),
M
a voxelwise data‐driven technique, was employed to compute local and global FCD maps. Then, short‐range and long‐range FCD values were calculated and group
d
analyses performed between patents with PNES and healthy controls. A correlation
te
analysis with clinical variables was also performed.
Ac ce p
Results: We found that patients with PNES showed abnormal FCD regions mainly in the frontal cortex, sensorimotor cortex, cingulate gyrus, insula and occipital cortex.
Seed‐voxel correlation analyses also showed disrupted functional connectivity between these regions. In addition, the occipital cortex FCD correlated with duration of disease.
Conclusion: The present results support the hypothesis that patients with PNES are associated with altered attention, sensorimotor and emotion systems. Furthermore, correlations between altered regions in the occipital cortex and duration of disease may reflect an adaptation in these patients for long‐term hypervigilance and 3
Page 3 of 37
increased response to external stimuli. This study adds new knowledge to our understanding of the pathophysiological mechanisms underlying PNES.
ip t
Keywords: long‐range FCD; short‐range FCD; resting‐state functional connectivity;
cr
sensorimotor system; psychogenic non‐epileptic seizures
Ac ce p
te
d
M
an
us
4
Page 4 of 37
1.1 Introduction Psychogenic non‐epileptic seizures (PNES) are paroxysmal behaviors with altered
ip t
movement, sensation, or experience, which resemble epileptic seizures, but are not accompanied by ictal epileptiform brain discharges (Baslet, 2011; Devinsky et al.,
cr
2011). PNES can be interpreted as an experiential or behavioral response to
us
emotional, psychological, or social distress (Reuber, 2008). Patients with PNES are often misdiagnosed and treated for epilepsy, which is detrimental because of the side
an
effects of antiepileptic drugs and the delay in proper treatment(Leis et al., 1992; Reuber et al., 2004). It has been observed that the diagnosis of PNES is usually
M
delayed for an average of 7 years (Reuber et al., 2002), significantly impacting
te
d
patients’ quality of life (Szaflarski et al., 2003b; Szaflarski et al., 2003a).
Etiologically, PNES is related to the dysfunction in processing of psychological or
Ac ce p
social distress (Baslet, 2011; Uliaszek et al., 2012), manifesting as an altered cognitive‐emotional attention system (Baslet, 2011). During episodes of PNES, sensorimotor and cognitive processes are affected and not properly integrated,
resulting in a range of involuntary behavioral patterns (Baslet, 2011). Findings from
recent neuroimaging studies have provided evidence to support this notion. Using
resting‐state functional magnetic resonance imaging (rs‐fMRI) technology, van der Kruijs et al. found dysfunctional connectivity between emotional, executive control, and sensorimotor networks in PNES (van der Kruijs et al., 2012). In addition, an EEG synchronization study revealed decreased prefrontal and parietal synchronization in 5
Page 5 of 37
PNES, reflecting dysfunction of fronto‐parietal networks (Knyazeva et al., 2011). Overall, these studies suggest aberrant functional connectivity in specific brain
ip t
networks, contributing to understanding the pathophysiological mechanism of PNES. Therefore, advanced neuroimaging technologies may aid the clinical diagnosis and
cr
treatment of PNES.
us
In the present study, we compared brain functional connectivity density (FCD)
an
between patients with PNES and healthy controls. Functional connectivity density mapping (FCDM) is a voxelwise data‐driven technique, recently proposed by Tomasi
M
and Volkow (Tomasi and Volkow, 2010). Simple voxelwise functional connectivity analyses only depict brain functional connectivity. In contrast, FCDM can locate highly
te
d
connected brain regions (functional hubs), and is broadly equivalent to the combination of simple voxelwise functional connectivity analyses and graph theory
Ac ce p
analysis. In addition, simple voxelwise functional connectivity analyses are
computationally demanding. However, FCDM is an ultrafast method and can compute local and global FCD maps with high spatial resolution (3‐mm isotropic), allowing identification of functional hubs (regions that are densely connected) with high sensitivity and discrimination among short‐range FCD hubs and long‐range FCD
hubs (Tomasi and Volkow, 2012b,a). Thus, the present study used FCDM analysis to investigate abnormal connectivity in PNES. We aimed to find regions exhibiting altered FCD in patients with PNES, and further investigate whether altered brain regions were related to the duration of disease. On the basis of previous findings and 6
Page 6 of 37
clinical symptoms of PNES, we hypothesized that regions with altered FCD in patients with PNES might be associated with attention, emotion, and sensorimotor systems.
ip t
1.2 Materials and Methods
cr
1.2.1 Participants
us
The participants were the same as in our previous study (Ding et al., 2013). A total of 20 patients with PNES (7 males, mean age: 19.65±7.56 years) and 20 healthy
an
volunteers (8 males, mean age: 21.85±1.70 years) from the Department of Neurology, West China Hospital, Chengdu, China were recruited. Patients with PNES were given
M
definitive diagnoses by experienced neurologists using clinical descriptions of symptoms and long‐term video/EEG monitoring, consistent with recent
te
d
recommendations (Benbadis et al., 2004; Devinsky et al., 2011). The inclusion criteria were: 1) at least one single typical episode recorded by video EEG, and EEG did not
Ac ce p
show any epileptiform discharge or ictal slowing; 2) patients had no history of neurological disease; 3) patients had no obvious abnormality in routine structural MRI examinations. The exclusion criteria were: 1) patients with neurological
comorbidity (e.g. epilepsy); 2) patients with malingering, or any psychiatric disorders
(e.g. mood and anxiety disorders, schizophrenia, and psychosis). Here, the diagnosis
of malingering or psychiatric disorders was determined by two attending psychiatrists using the Structured Clinical Interview for DSM‐IV (SCID)‐Patients Version and their scores on the Hamilton Anxiety Rating Scale and Hamilton Depression Rating Scale. Only patients with a diagnosis of definite PNES were included in the study. Four out 7
Page 7 of 37
of 20 patients were taking antiepileptic drugs before the diagnosis of PNES. All drugs were discontinued at least 2 weeks prior to MRI examination. Demographic and
ip t
Clinical Characteristics of the patients with PNES are shown in Table 1. To increase the homogeneity of the patient group, two patients with very long duration of disease
cr
(approximately 18 years) were excluded from the study, resulting in a final analysis of
us
18 patients with PNES and 20 healthy controls. The control subjects had no neurologic/psychiatric disorders, evaluated using the SCID‐Non‐Patient Version, and
an
had not taken any psychotropic medication within the past 6 months. This study was approved by the Local Ethics Committee of West China Hospital, and written
M
informed consent was obtained from each subject before experimentation.
te
1.2.2 Data Acquisition
d
All imaging data were collected on a 3T Siemens Trio system (Erlangen, Germany) at
Ac ce p
the MR Research Center of West China Hospital, Chengdu, China. During data
acquisition, subjects were instructed to relax with their eyes closed, and to keep their heads still. Foam padding and earplugs were used to reduce head motion and scanner noise. Functional images were acquired using a single‐shot, gradient‐recalled echo planar imaging sequence for a total of 205 volumes (repetition time [TR]/echo
time [TE] = 2000/30 ms; flip angle = 90°; field of view = 240×240 mm2; in‐plane matrix = 64×64; voxel size = 3.75×3.75×5 mm3, no slice gap; 30 axial slices). Additionally, high‐resolution T1‐weighted anatomical images were also acquired using a magnetization‐prepared rapid gradient‐echo sequence (TR/TE = 20/3.69 ms; flip angle 8
Page 8 of 37
= 12°; field of view = 250×250 mm2; in‐plane matrix = 320×320; voxel size = 0.78×0.78×1 mm3, no slice gap, 128 sagittal slices) for each subject.
ip t
1.2.3 Data Preprocessing
cr
Functional images preprocessing was performed using the Statistical Parametric
us
Mapping software (SPM8, http://www.fil.ion.ucl.ac.uk/spm). The first five volumes were not analyzed to allow for signal equilibration effects. The remaining 200
an
consecutive volumes were corrected for temporal differences and head motion, spatially normalized to the Montreal Neurological Institute (MNI) echo‐planar
M
imaging template and resampled to 3‐mm cubic voxels. There were no subjects with movement greater than 1.5 mm translation or 1.5° rotation. Recent studies have
te
d
shown that functional connectivity analysis is sensitive to gross head motion effects (Power et al., 2012; Van Dijk et al., 2012); therefore, we further evaluated the mean
Ac ce p
absolute displacement of each brain volume as compared with the previous volume(Van Dijk et al., 2012). The largest mean displacement (MD) of all subjects was
less than 0.2 mm, and there was no significant difference in mean displacement between patients with PNES and healthy controls (MD: 0.0838±0.0319 for PNES and 0.0869±0.0316 for healthy controls; p = 0.7568 ) using a two‐sample two‐tailed t test. For each subject, we extracted all voxels’ time series data from brain gray matter (not including the cerebellum). In the present study, we only investigated cerebral
alterations in patients with PNES, thus the cerebellum was excluded. Next, the time series of each voxel was corrected using a linear regression process to remove several 9
Page 9 of 37
spurious sources of variances, including six head motion parameters and averaged signals from ventricles and white matter. The residuals of these regressions were
ip t
temporally band‐pass filtered (0.01‐0.08 Hz) to reduce low‐frequency drift (Foerster et al., 2005) and high‐frequency noise related to respiratory and other physiological
cr
processes (Cordes et al., 2001).
us
1.2.4 Functional Connectivity Density
an
The preprocessed image data underwent FCD mapping (Tomasi and Volkow, 2010) to compute the strength of the local FCD (lFCD) and global FCD (gFCD), a detailed
M
description of the computation of lFCD and gFCD is given in Tomasi and Volkow (2010). Briefly, we first defined the number of functional connections, ki , which was
te
d
calculated using Pearson correlation coefficients between the time series of voxel i and the other voxels using a given threshold T . Here, the correlation coefficient
Ac ce p
threshold T was set to 0.44, corresponding to p < 0.05 , family‐wise error (FWE)‐corrected. This correlation coefficient threshold was used to reduce the
chance of false positive connections across all subjects.
According to the definition of Tomasi and Volkow (2010), the lFCD at a given voxel i is the local ki between i and its neighbor voxels using a three‐dimensional searching algorithm developed in Interactive Data Language (IDL). When all the neighbors of voxel i are detected, the local ki can be calculated. Then the same steps are repeated for the next voxel. After obtaining all voxels’ local functional 10
Page 10 of 37
connections, the lFCD mapping of each subject was obtained.
ip t
For a given voxel i , the gFCD was defined as the global functional connections, ki , between i and all other voxels in the brain. Two voxels were considered
cr
functionally connected if the correlation coefficient was larger than 0.44. The
us
calculation was repeated for all voxels in the brain. A simple approach, based on a parallel algorithm, was developed in C‐language to speed up calculation of the gFCD
an
by taking advantage of multiprocessor computer architectures.
1.2.5 Short‐ and Long‐range FCD
M
The lFCD predominantly reflects functional connectivity of the local cluster, so the
te
d
short‐range FCD is equated to the lFCD. As the gFCD included both local and distal functional connections, the long‐range FCD is defined as gFCD‐lFCD (Tomasi and
Ac ce p
Volkow, 2012b,a). To minimize differences in functional anatomy of the brain across
subjects, short‐ and long‐range FCD maps were spatially smoothed by convolution
with an isotropic Gaussian kernel (FWHM=8mm) in SPM8 (Tomasi and Volkow, 2012b,a). For each subject, short‐ and long‐range FCD distributions were further scaled by the average strength in the whole brain respectively to reduce individual
overall differences in the strength of FCD. 1.2.6 Statistical analysis One‐way analysis of covariance (ANCOVA) with three covariates (age, gender and 11
Page 11 of 37
mean displacement) was implemented in SPM8 to compare group differences in short‐ and long‐range FCD respectively. The statistical significance of group differences was set at t > 2.4377 (individual voxel threshold p < 0.01 , df = (1,36) )
ip t
and minimum cluster size of 59 voxels, corresponding to a corrected p < 0.05 . This
cr
correction was confined within the brain gray matter (not including the cerebellum)
us
and was determined by the Monte Carlo simulations performed by the AlphaSim program in the REST toolkit (http://sourceforge.net/projects/resting‐fmri).
an
Further, clusters with significant differences were evaluated using regions of interest
M
(ROI) analysis. Each ROI was a 9‐mm isotropic cube including 27 voxels (0.73cm3), centered at the Montreal Neurological Institute coordinates of the local maxima. A
te
d
correlation analysis was then performed between these ROIs and duration of disease. Using standard seed‐voxel correlation analyses, we further examined resting‐state
Ac ce p
functional connectivity (RSFC) maps through Pearson correlation coefficients between the averaged time series of the abnormal ROIs and those of other voxels in the brain gray matter, respectively. The correlation coefficient maps were converted
into z maps by Fisher’s r ‐to‐ z transformation to improve normality, and the z maps were spatially smoothed (FWHM=8mm) like the FCD maps. Group comparisons were performed using one‐way ANCOVA, and the statistical threshold was set at
p < 0.05 (corrected for multiple comparisons using the AlphaSim program).
1.3 Results 12
Page 12 of 37
1.3.1 Short‐ and long‐range FCD distributions Fig. 1A and B show the average distribution of short‐range FCD in patients with PNES
ip t
and healthy controls respectively. For both patients with PNES and healthy controls, the high short‐range FCD was mainly distributed in the bilateral precuneus, cuneus,
cr
occipital cortex, parietal cortex, postcentral gyrus, and middle temporal gyrus. Fig. 2A
us
and B show the average distribution of long‐range FCD in patients with PNES and healthy controls, respectively. The high long‐range FCD was also distributed in the
an
bilateral posterior cingulate/precuneus, median cingulate gyrus, parietal cortex, angular, middle temporal gyrus and dorsolateral prefrontal cortex. Regions with high
M
FCD indicate that they might play important roles in brain networks, namely hubs. The FCD hubs found in the present study are consistent with previous studies
te
d
(Buckner et al., 2009; Tomasi and Volkow, 2012b).
Ac ce p
1.3.2 Group Comparisons
We found that patients with PNES showed increased short‐range FCD mainly in the
left middle frontal gyrus, superior frontal gyrus, medial part of the superior frontal
gyrus, anterior cingulate gyrus, supplementary motor area, and bilateral median cingulate gyrus, and decreased short‐range FCD mainly in the right middle occipital
gyrus compared with healthy controls (Fig. 1C, p < 0.05 , AlphaSim corrected). The results of ROI analysis are shown in Table 2. Compared with healthy controls, patients with PNES showed increased long‐range 13
Page 13 of 37
FCD mainly in the bilateral calcarine fissure, lingual gyrus, supplementary motor area, and the right superior temporal gyrus, insula, precentral and postcentral gyrus, and
ip t
the left paracentral lobule; and decreased long‐range FCD mainly in the right medial prefrontal cortex, middle frontal gyrus, triangular and opercular parts of the inferior
cr
frontal gyrus, superior frontal gyrus, medial part of the superior frontal gyrus,
us
supramarginal gyrus and inferior parietal gyrus (Fig. 2C, p < 0.05 , AlphaSim corrected). The results of the ROI analysis are shown in Table 3.
an
1.3.3 RSFC Networks
M
To further explore the hypo/hyper connectivity between brain regions and abnormal short‐ and long‐range FCD regions in patients with PNES, we examined functional
te
d
connectivity networks in patients and healthy controls.
Ac ce p
The RSFC maps for abnormal FCD regions belonging to the same brain system, i.e. the left frontal, cingulate, visual, sensorimotor, and right frontal cortex, were combined using a conjunction analysis (Fox et al., 2005; Zhang et al., 2011). For the RSFC
conjunctions of the left middle, superior, and medial part of the superior frontal gyrus seeds, we found increased functional connectivity among those seeds and in
the anterior cingulate gyrus and right median cingulate gyrus in patients with PNES (Fig. 3A, p < 0.05 , AlphaSim corrected). For the RSFC conjunctions of the anterior cingulate gyrus and bilateral median cingulate gyrus seeds, we observed that the patients with PNES exhibited increased functional connectivity mainly between each 14
Page 14 of 37
other and in the frontal cortex, supplementary motor area and precuneus (Fig. 3B,
p < 0.05 , AlphaSim corrected). For the RSFC conjunctions of the bilateral
ip t
supplementary motor area, right precentral and postcentral gyrus, and left paracentral lobule seeds, we found increased functional connectivity mainly between
cr
each other and in the bilateral precuneus, median cingulate gyrus, superior temporal
us
cortex, and right pole part of the superior temporal cortex (Fig. 3D, p < 0.05 , AlphaSim corrected) in the patients with PNES. For the RSFC conjunctions of the
an
bilateral calcarine fissure and lingual gyrus seeds, we found increased functional connectivity between each other and in the bilateral cuneus, postcentral gyrus and
M
left precuneus, and decreased functional connectivity in the bilateral fusiform gyrus,
te
patients with PNES.
d
middle and superior occipital cortex (Fig. 3E, p < 0.05 , AlphaSim corrected) in the
Ac ce p
For the RSFC conjunctions of the right medial prefrontal cortex, middle frontal gyrus,
triangular and opercular parts of the inferior frontal gyrus, superior and medial parts
of the superior frontal gyrus seeds, we found common lower functional connectivity in the right inferior parietal, supramarginal and angular gyrus in patients with PNES (Fig. 3F, p < 0.05 , AlphaSim corrected).
The RSFC maps of the right insula exhibited increased functional connectivity in the bilateral superior temporal cortex, supplementary motor area, rolandic area and left supramarginal gyrus (Fig. 3C, p < 0.05 , AlphaSim corrected). 15
Page 15 of 37
1.3.4 Correlation with Duration of Disease
ip t
Spearman correlations were calculated between the abnormal regions and duration of disease after removing potential outliers (Schwarzkopf et al., 2012). As seen in Fig.
cr
4, three regions with increased long‐range FCD in patients with PNES were
us
significantly positively correlated with disease duration ( p < 0.05 ), including the bilateral lingual gyrus, and the right calcarine fissure. For short‐range FCD, no regions
an
were found to correlate with disease duration.
M
1.4 Discussion
The present study investigated map changes in brain functional connectivity,
te
d
including short‐ and long‐range FCD, in patients with PNES using a voxelwise data‐driven approach. The regions with abnormal functional connectivity density in
Ac ce p
patients with PNES were mainly involved in the frontal cortex, sensorimotor cortex, cingulate gyrus, insula, and occipital cortex, supporting the hypothesis that PNES are
associated with altered attention, sensorimotor and emotion systems. Furthermore,
some altered regions in the occipital cortex were correlated with duration of disease. This study provides new insights into our understanding of the pathophysiological mechanisms underlying PNES. 1.4.1 Altered Short‐range FCD Compared with healthy controls, regions with increased short‐range FCD in patients 16
Page 16 of 37
with PNES were mainly involved in the left frontal cortex, anterior cingulate gyrus, and the bilateral median cingulate gyrus (Table 2). The frontal cortex is considered
ip t
the emotional control center, and the left frontal cortex is responsible for language skills, problem solving ability, impulse control, and judgment (Miller and Cummings,
cr
1999). The cingulate cortex is engaged in self‐control of cognitive processing and
us
attention, novelty detection, and task implementation (Bush et al., 2000; Dosenbach et al., 2006; Paus et al., 1993; Paus, 2001). Previous studies have found that patients
an
with PNES show hyper responses to novel stimuli, and fail to neglect irrelevant stimuli with social emotion (Gene‐Cos et al., 2005; Pouretemad et al., 1998).
M
Moreover, patients with PNES experience more stress, are more likely to deny their experiences of psychological stress, and to choose avoidance oriented coping
te
d
strategies, such as behavioral efforts to avoid conflicts or stress (Bakvis et al., 2011; Frances et al., 1999; Tojek et al., 2000). The findings of increased short‐range FCD in
Ac ce p
the left frontal and cingulate cortex in patients with PNES suggest that the function of
these regions may be impaired. Furthermore, the RSFC maps based on seeds in the left frontal and cingulate cortex exhibited hyperconnectivity within each system and
also between each other. Therefore, our findings support an altered
cognitive‐emotional attention system in patients the PNES (Baslet, 2011).
1.4.2 Altered Long‐range FCD Patients with PNES showed increased long‐range FCD mainly in the insula, sensorimotor (including bilateral supplementary motor area, right precentral and 17
Page 17 of 37
postcentral gyrus) and occipital cortex (including bilateral calcarine fissure and lingual gyrus) (Table 3). The insula is an important multisensory integration area that
ip t
mediates interpretation of sensory information from the body, related to emotion regulation, visceral sensory perception, and self‐awareness (Craig, 2002; Pollatos et
cr
al., 2007). Altered long‐range FCD found in the insula supports that PNES is involved
us
in impaired function of emotion and cognition (Reuber et al., 2004).
an
PNES are episodes of altered movement and sensation (Brown et al., 2011). During episodes of PNES, sensorimotor and cognitive processes are affected and not
M
properly integrated, resulting in a range of involuntary behavioral patterns (Baslet, 2011). For most patients, these involuntary behavioral patterns are avoidant coping
te
d
strategies used to deal with stressors, to keep them from experiencing stressful events or from memories of those events (Bakvis et al., 2011; Magaudda et al., 2011).
Ac ce p
A recent rs‐fMRI study reported that patients with PNES showed abnormal functional connectivity in the sensorimotor cortex (van der Kruijs et al., 2012). In addition, Labate et al. (2012) found abnormal cortical atrophy of motor regions. Combining
these previous findings and the known symptoms of PNES, our results of altered
long‐range FCD in the sensorimotor cortex provide further evidence that the sensorimotor cortex plays an important role in the pathogenesis of PNES. Moreover, seed‐voxel correlation analysis showed hyperconnectivity between the insula and supplementary motor area, and between the sensorimotor and cingulate cortex, supporting dysfunctional connectivity between emotional and sensorimotor 18
Page 18 of 37
networks in PNES (van der Kruijs et al., 2012). The present findings also provide neuroimaging evidence for understanding the pathology of PNES, as PNES are
in the processing of psychological or social distress (Baslet, 2011).
cr
ip t
associated with involuntary changes in movement and sensation due to a dysfunction
us
The calcarine fissure and lingual gyrus are thought to be related to visual memory and face memory (Kapur et al., 1995). In an experimental study, patients with PNES
an
were found to show high attentional bias to angry faces, and were likely to take avoidant strategies, i.e. related motor responses, to deal with the angry faces (Bakvis
M
et al., 2011). In the present study, we found that patients with PNES exhibited altered long‐range FCD in the occipital cortex, and the alterations correlated with duration of
te
d
disease, which might reflect an adaptation in patients with PNES for long‐term hypervigilance and increased response to external stimuli.
Ac ce p
In addition, we found regions with decreased long‐range FCD mainly in the right frontal cortex in patients with PNES. The frontal cortex plays an important role in maintaining long‐term memories, which are involved in emotions derived from the
limbic system (Buckner and Petersen, 1996). The right frontal cortex is associated with social cognition and emotions, especially for negative emotions (Miller and Cummings, 1999). A previous study demonstrated that cognitive integrative functions for dealing with social stress and memory are impaired in patients with PNES (Bakvis
et al., 2010). Our findings further support this finding. For the seed‐voxel correlation 19
Page 19 of 37
analyses, patients with PNES showed decreased functional connectivity between the right frontal and parietal cortex (including the right inferior parietal, supramarginal,
ip t
and angular gyrus). The fronto‐parietal network is associated with attentional control, and is activated when attention is directed to external stimuli in cognitive tasks (Fox
cr
et al., 2005; Seeley et al., 2007; Sridharan et al., 2008). Therefore, decreased
us
functional connectivity between the right frontal and parietal cortex indicates that the fronto‐parietal network may be impaired, providing evidence for dysfunction of
an
cognitive attention and executive control in patients with PNES (Baslet, 2011).
M
Some methodological limitations should be mentioned in this study. First, as our sample size was modest, a relatively weak correction strategy (AlphaSim program)
te
d
was used for multiple comparisons in this study. Thus, the present results require replication using larger sample sizes. Second, recent studies have shown significant
Ac ce p
effects of head motion on functional connectivity analysis (Power et al., 2012; Van Dijk et al., 2012). We examined the mean displacement of head motion and found no significant between‐group difference. Nevertheless, we partialed out the effect of
mean displacement of head motion in our statistical comparisons to reduce the effect of head motion. Third, the correlations between disease duration and regions
with altered FCD were not corrected for multiple comparisons. Finally, as this was an exploratory study, we only selected a single threshold to calculate FCD maps. In addition, a range of thresholds can also be used to test the stability of the results. 20
Page 20 of 37
1.5 Conclusion In conclusion, we investigated the map changes in brain functional connectivity in
ip t
patients with PNES. The abnormal FCD regions and functional connectivity based on these regions in patients with PNES were mainly associated with attention,
cr
sensorimotor, and emotion systems, supporting our hypothesis. Furthermore, the
us
altered regions were also involved in the occipital cortex, and some alterations correlated with duration of disease, which might reflect an adaptation in patients
an
with PNES for long‐term hypervigilance and increased response to external stimuli. This study improves our understanding of the pathophysiological mechanisms
Ac ce p
te
d
M
underlying PNES.
21
Page 21 of 37
Acknowledgments: This work was supported by the 973 Project (2012CB517901); the Natural Science Foundation of China (61035006, 61125304); and the Academic
Ac ce p
te
d
M
an
us
cr
Author Disclosure Statement: No competing financial interests exist.
ip t
new artist Ministry of Education doctoral post graduate grant (A03003023901004).
22
Page 22 of 37
References
Ac ce p
te
d
M
an
us
cr
ip t
Bakvis, P., Spinhoven, P., Putman, P., Zitman, F. G., Roelofs, K., 2010. The effect of stress induction on working memory in patients with psychogenic nonepileptic seizures. Epilepsy Behav 19, 448-454. Bakvis, P., Spinhoven, P., Zitman, F. G., Roelofs, K., 2011. Automatic avoidance tendencies in patients with psychogenic non-epileptic seizures. Seizure 20, 628-634. Baslet, G., 2011. Psychogenic non-epileptic seizures: a model of their pathogenic mechanism. Seizure 20, 1-13. Benbadis, S. R., Siegrist, K., Tatum, W. O., Heriaud, L., Anthony, K., 2004. Short-term outpatient EEG video with induction in the diagnosis of psychogenic seizures. Neurology 63, 1728-1730. Brown, R. J., Syed, T. U., Benbadis, S., LaFrance, W. C., Jr., Reuber, M., 2011. Psychogenic nonepileptic seizures. Epilepsy Behav 22, 85-93. Buckner, R. L., Petersen, S. E., 1996. What does neuroimaging tell us about the role of prefrontal cortex in memory retrieval? Seminars in the Neurosci 8, 47-55. Buckner, R. L., Sepulcre, J., Talukdar, T., Krienen, F. M., Liu, H., Hedden, T., Andrews-Hanna, J. R., Sperling, R. A., Johnson, K. A., 2009. Cortical hubs revealed by intrinsic functional connectivity: mapping, assessment of stability, and relation to Alzheimer's disease. J Neurosci 29, 1860-1873. Bush, G., Luu, P., Posner, M. I., 2000. Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn Sci 4, 215-222. Cordes, D., Haughton, V. M., Arfanakis, K., Carew, J. D., Turski, P. A., Moritz, C. H., Quigley, M. A., Meyerand, M. E., 2001. Frequencies contributing to functional connectivity in the cerebral cortex in "resting-state" data. AJNR Am J Neuroradiol 22, 1326-1333. Craig, A. D., 2002. How do you feel? Interoception: the sense of the physiological condition of the body. Nat Rev Neurosci 3, 655-666. Devinsky, O., Gazzola, D., LaFrance, W. C., Jr., 2011. Differentiating between nonepileptic and epileptic seizures. Nat Rev Neurol 7, 210-220. Ding, J. R., An, D., Liao, W., Li, J., Wu, G. R., Xu, Q., Long, Z., Gong, Q., Zhou, D., Sporns, O., Chen, H., 2013. Altered functional and structural connectivity networks in psychogenic non-epileptic seizures. PLoS One 8, e63850. Dosenbach, N. U., Visscher, K. M., Palmer, E. D., Miezin, F. M., Wenger, K. K., Kang, H. C., Burgund, E. D., Grimes, A. L., Schlaggar, B. L., Petersen, S. E., 2006. A core system for the implementation of task sets. Neuron 50, 799-812. Foerster, B. U., Tomasi, D., Caparelli, E. C., 2005. Magnetic field shift due to mechanical vibration in functional magnetic resonance imaging. Magn Reson Med 54, 1261-1267. Fox, M. D., Snyder, A. Z., Vincent, J. L., Corbetta, M., Van Essen, D. C., Raichle, M. E., 2005. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci U S A 102, 9673-9678. Frances, P. L., Baker, G. A., Appleton, P. L., 1999. Stress and avoidance in Pseudoseizures: testing the assumptions. Epilepsy Res 34, 241-249. Gene-Cos, N., Pottinger, R., Barrett, G., Trimble, M. R., Ring, H. A., 2005. A comparative study of mismatch negativity (MMN) in epilepsy and non-epileptic seizures. Epileptic Disord 7, 363-372. Kapur, N., Friston, K. J., Young, A., Frith, C. D., Frackowiak, R. S., 1995. Activation of human hippocampal formation during memory for faces: a PET study. Cortex 31, 99-108. Knyazeva, M. G., Jalili, M., Frackowiak, R. S., Rossetti, A. O., 2011. Psychogenic seizures and frontal disconnection: EEG synchronisation study. J Neurol Neurosurg Psychiatry 82, 505-511. Labate, A., Cerasa, A., Mula, M., Mumoli, L., Gioia, M. C., Aguglia, U., Quattrone, A., Gambardella, A., 2012. Neuroanatomic correlates of psychogenic nonepileptic seizures: A cortical thickness and VBM study. Epilepsia 53, 377-385. Leis, A. A., Ross, M. A., Summers, A. K., 1992. Psychogenic seizures: ictal 23
Page 23 of 37
Ac ce p
te
d
M
an
us
cr
ip t
characteristics and diagnostic pitfalls. Neurology 42, 95-99. Magaudda, A., Gugliotta, S. C., Tallarico, R., Buccheri, T., Alfa, R., Lagana, A., 2011. Identification of three distinct groups of patients with both epilepsy and psychogenic nonepileptic seizures. Epilepsy Behav 22, 318-323. Miller, B. L., Cummings, J. L., 1999. The human frontal lobes : functions and disorders. Guilford Press: New York. Paus, T., Petrides, M., Evans, A. C., Meyer, E., 1993. Role of the human anterior cingulate cortex in the control of oculomotor, manual, and speech responses: a positron emission tomography study. J Neurophysiol 70, 453-469. Paus, T., 2001. Primate anterior cingulate cortex: where motor control, drive and cognition interface. Nat Rev Neurosci 2, 417-424. Pollatos, O., Gramann, K., Schandry, R., 2007. Neural systems connecting interoceptive awareness and feelings. Hum Brain Mapp 28, 9-18. Pouretemad, H. R., Thompson, P. J., Fenwick, P. B., 1998. Impaired sensorimotor gating in patients with non-epileptic seizures. Epilepsy Res 31, 1-12. Power, J. D., Barnes, K. A., Snyder, A. Z., Schlaggar, B. L., Petersen, S. E., 2012. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. Neuroimage 59, 2142-2154. Reuber, M., Fernandez, G., Bauer, J., Helmstaedter, C., Elger, C. E., 2002. Diagnostic delay in psychogenic nonepileptic seizures. Neurology 58, 493-495. Reuber, M., Baker, G. A., Gill, R., Smith, D. F., Chadwick, D. W., 2004. Failure to recognize psychogenic nonepileptic seizures may cause death. Neurology 62, 834-835. Reuber, M., 2008. Psychogenic nonepileptic seizures: answers and questions. Epilepsy Behav 12, 622-635. Schwarzkopf, D. S., De Haas, B., Rees, G., 2012. Better ways to improve standards in brain-behavior correlation analysis. Front Hum Neurosci 6, 200. Seeley, W. W., Menon, V., Schatzberg, A. F., Keller, J., Glover, G. H., Kenna, H., Reiss, A. L., Greicius, M. D., 2007. Dissociable intrinsic connectivity networks for salience processing and executive control. J Neurosci 27, 2349-2356. Sridharan, D., Levitin, D. J., Menon, V., 2008. A critical role for the right fronto-insular cortex in switching between central-executive and default-mode networks. Proc Natl Acad Sci U S A 105, 12569-12574. Szaflarski, J. P., Hughes, C., Szaflarski, M., Ficker, D. M., Cahill, W. T., Li, M., Privitera, M. D., 2003a. Quality of life in psychogenic nonepileptic seizures. Epilepsia 44, 236-242. Szaflarski, J. P., Szaflarski, M., Hughes, C., Ficker, D. M., Cahill, W. T., Privitera, M. D., 2003b. Psychopathology and quality of life: psychogenic non-epileptic seizures versus epilepsy. Med Sci Monit 9, CR113-118. Tojek, T. M., Lumley, M., Barkley, G., Mahr, G., Thomas, A., 2000. Stress and other psychosocial characteristics of patients with psychogenic nonepileptic seizures. Psychosomatics 41, 221-226. Tomasi, D., Volkow, N. D., 2010. Functional connectivity density mapping. Proc Natl Acad Sci U S A 107, 9885-9890. Tomasi, D., Volkow, N. D., 2012a. Aging and functional brain networks. Mol Psychiatry 17, 471, 549-458. Tomasi, D., Volkow, N. D., 2012b. Abnormal functional connectivity in children with attention-deficit/hyperactivity disorder. Biol Psychiatry 71, 443-450. Uliaszek, A. A., Prensky, E., Baslet, G., 2012. Emotion regulation profiles in psychogenic non-epileptic seizures. Epilepsy Behav 23, 364-369. van der Kruijs, S. J., Bodde, N. M., Vaessen, M. J., Lazeron, R. H., Vonck, K., Boon, P., Hofman, P. A., Backes, W. H., Aldenkamp, A. P., Jansen, J. F., 2012. Functional connectivity of dissociation in patients with psychogenic non-epileptic seizures. J Neurol Neurosurg Psychiatry 83, 239-247. Van Dijk, K. R., Sabuncu, M. R., Buckner, R. L., 2012. The influence of head motion on intrinsic functional connectivity MRI. Neuroimage 59, 431-438. Zhang, Z., Liao, W., Zuo, X. N., Wang, Z., Yuan, C., Jiao, Q., Chen, H., Biswal, B. B., Lu, G., Liu, Y., 2011. Resting-state brain organization revealed by functional 24
Page 24 of 37
covariance networks. PLoS One 6, e28817.
Ac ce p
te
d
M
an
us
cr
ip t
25
Page 25 of 37
Table 1 Demographic and Clinical Characteristics of patients with PNES Previous Patient
Age
Gender
Duration
Type of symptoms
1
34y
F
18y
Unresponsiveness/eye closure
2
17y
F
2y
Unresponsiveness/eye
ip t
Treatment* None
cr
None
rigidity
us
closure/hyperventilation/body
38y
F
18y
Unresponsiveness/hypermotor EX
4
23y
F
8y
Unresponsiveness/eye
an
3
None None
M
closure/hyperventilation/body
F
6
20y
M
7
14y
F
8
17y
F
1m
Unresponsiveness/eye closure
None
2y
Unresponsiveness
VPA
2m
Unresponsiveness
None
2m
Unresponsiveness/eye
None
te
13y
Ac ce p
5
d
rigidity
closure/body rigidity/trembling EX
9
14y
M
5m
Unresponsiveness/vocalization
VPA
10
16y
F
2y
Unresponsiveness/hyperventilation
None
/hypermotor EX 11
17y
F
4m
Unresponsiveness/hyperventilation
None
12
21y
F
8m
Unresponsiveness/hypermotor EX
None
13
21y
M
1m
Eye closure/hyperventilation/body None 26
Page 26 of 37
rigidity 14
13y
M
1y
Unresponsiveness/eye
None
15
13y
F
1y
ip t
closure/hyperventilation Unresponsiveness/eye
None
35y
M
15d
Unresponsiveness/eye
us
16
cr
closure/hyperventilation
None
closure/hypermotor EX F
3y
Unresponsiveness
18
20y
M
2y
Unresponsiveness/hypermotor EX
VPA
19
13y
M
1m
Unresponsiveness
None
20
18y
F
7m
Unresponsiveness/eye closure
None
CBZ
d
M
an
16y
te
17
*: All drugs were discontinued at least 2 weeks before MRI examination.
Ac ce p
Abbreviations: F: female; M: male; d: day; m: month; y: year; Hypermotor EX: Hypermotor movements of the extremities; Trembling EX: Trembling of the
extremities; VPA: valproate; CBZ: carbamazepine.
27
Page 27 of 37
Table 2 Statistical Significance (t Score; one‐way ANOVA) of Short‐Range FCD Differences between patients with PNES and HCs BA
x(mm)
y(mm)
z(mm)
Short‐range FCD increased regions
PNES>HCs(t)
ip t
Anatomical region
46
‐27
51
24
SFG_L
32
‐12
27
39
3.04
SFG_Med_L
32
‐3
21
42
ACG_L
32
‐9
33
27
4.01
SMA_L
6
‐3
MCG_R
—
3
MCG_L
—
us
18
45
3.09
‐24
48
4.45
‐24
48
3.89
‐81
27
‐4.35
M
an
3.58
d
0
3.82
cr
MFG_L
MOG_R
19
42
Ac ce p
te
Short‐range FCD decreased regions
Statistical values averaged in 9‐mm isotropic (cube; 27 voxels) regions of interest centered at the Montreal Neurological Institute coordinates (x, y, z) of the local maxima.
FCD, functional connectivity density; ANOVA, analysis of variance; PNES, psychogenic
non‐epileptic seizures; HCs, healthy controls; BA, Brodmann’s area; L, Left; R, Right; MFG, middle frontal gyrus; SFG, superior frontal gyrus; SFG_Med, medial part of superior frontal gyrus; ACG, anterior cingulate gyrus; SMA, supplementary motor area; MCG, median cingulate gyrus; MOG, middle occipital gyrus. 28
Page 28 of 37
Table 3 Statistical Significance (t Score; one‐way ANOVA) of Long‐Range FCD Differences between patients with PNES and HCs BA
x(mm)
y(mm)
z(mm)
Long‐range FCD increased regions
PNES>HCs(t)
ip t
Anatomical region
17
0
‐84
0
CAL_R
17
6
‐75
6
3.21
LING_R
17
6
‐69
6
4.02
LING_L
17
0
‐72
6
STG_R
22
48
INS_R
48
45
PCL_L
4
‐6
PreCG_R
6
SMA_R
us
‐9
‐6
3.46
‐9
0
3.63
‐28
75
3.07
‐21
75
3.72
d
M
an
3.26
4
24
‐33
72
2.90
—
‐3
‐12
66
3.02
6
6
‐15
78
3.11
Ac ce p
SMA_L
te
PoCG_R
21
3.99
cr
CAL_L
Long‐range FCD decreased regions MFG_Orb_R
46
39
57
‐6
‐3.92
MFG_R
10
36
57
0
‐3.37
IFG_Tri_R
48
51
21
21
‐3.10
IFG_Oper_R
44
51
21
33
‐3.36
SMG_R
40
57
‐33
42
‐3.45
IPG_R
40
39
‐45
39
‐3.38
29
Page 29 of 37
SFG_R
8
24
15
66
‐3.58
SFG_Med_R
8
12
30
63
‐3.21
ip t
Statistical values averaged in 9‐mm isotropic (cube; 27 voxels) regions of interest
cr
centered at the Montreal Neurological Institute coordinates (x, y, z) of the local
us
maxima.
FCD, functional connectivity density; ANOVA, analysis of variance; PNES, psychogenic
an
non‐epileptic seizures; HCs, healthy controls; BA, Brodmann’s area; L, Left; R, Right; CAL, calcarine fissure; LING, lingual gyrus; STG, superior temporal gyrus; INS, insula;
M
PCL, paracentral lobule; PreCG, precentral gyrus; PoCG, postcentral gyrus; SMA,
d
supplementary motor area; MFG_Orb, medial prefrontal cortex; MFG, middle frontal
te
gyrus; IFG_Tri, triangular part of inferior frontal gyrus; IFG_Oper, opercular part of inferior frontal gyrus; SMG, supramarginal gyrus; IPG, inferior parietal gyrus; SFG,
Ac ce p
superior frontal gyrus; SFG_Med, medial part of superior frontal gyrus.
30
Page 30 of 37
Captions to Figure Fig. 1 Distribution of short‐range FCD in the brain for 18 patients with PNES and 20
ip t
HCs and statistical differences ( t scores) between the groups. One‐way analysis of covariance with three covariates (age, gender, and mean motion) was used to
cr
contrast short‐range FCD maps across groups. The statistical significance of group
us
differences was set at t > 2.4377 (individual voxel threshold p < 0.01 , df = (1,36) ) and minimum cluster size of 59 voxels, corresponding to an AlphaSim corrected
an
p < 0.05 . The left hemisphere is on the left. df, degree of freedom; PNES, psychogenic non‐epileptic seizures; HCs, healthy controls.
M
Fig. 2 Distribution of long‐range FCD in the brain for 18 patients with PNES and 20
te
d
HCs and the statistical differences ( t scores) between the two groups. One‐way analysis of covariance with three covariates (age, gender, and mean motion) was
Ac ce p
used to contrast long‐range FCD maps across groups. The statistical significance of group
differences
was
set
at
t > 2.4377
(individual voxel threshold
p < 0.01 , df = (1,36) ) and minimum cluster size of 59 voxels, corresponding to an
AlphaSim corrected p < 0.05 . The left hemisphere is on the left. df, degree of freedom; PNES, psychogenic non‐epileptic seizures; HCs, healthy controls.
Fig. 3 Statistical significance of RSFC maps for the abnormal FCD regions between patients with PNES and HCs ( p < 0.05 , AlphaSim corrected). Group differences in conjunction analyses of the RSFC maps with seeds belonging to the left frontal cortex 31
Page 31 of 37
(A), cingulate cortex (B), right insula (C), sensorimotor cortex (D), visual cortex (E) and right frontal cortex (F), respectively. Each result is displayed on three ‘brains’ shown
ip t
from the right side (top left image of each group), from behind (top right image of each group), and from above (bottom left image of each group). Warm and cool
cr
colors indicate RSFC increases and decreases, respectively, in the patients with PNES.
us
One‐way analysis of covariance with three covariates (age, gender, and mean motion) was used to compare group differences. RSFC, resting‐state functional connectivity;
an
PNES, psychogenic non‐epileptic seizures; HCs, healthy controls.
M
Fig. 4 Correlations between regions with altered long‐range FCD and duration of disease in patients with PNES ( p < 0.05 ). Spearman correlations were calculated over
te
d
the data after removing outliers marked by circles. Cubic ROI volume: 27 voxels; ROI center coordinates shown in Table 3. CAL, calcarine fissure; LING, lingual gyrus; L, Left;
Ac ce p
R, Right.
32
Page 32 of 37
Highlights We examine changes of brain functional connectivity in PNES patients. Abnormal regions in PNES are related to attention, sensorimotor and emotion systems.
ip t
Altered regions in the occipital cortex are correlated with duration of disease.
Ac ce p
te
d
M
an
us
cr
33
Page 33 of 37
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 34 of 37
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 35 of 37
Ac ce p
te
d
M
an
us
cr
ip t
Figure(s)
Page 36 of 37
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 37 of 37
S0920-1211(14)00137-5 http://dx.doi.org/doi:10.1016/j.eplepsyres.2014.05.006 EPIRES 5156
To appear in:
Epilepsy Research
Received date: Revised date: Accepted date:
14-11-2013 14-4-2014 6-5-2014
Please cite this article as: Ding, J.-R., An, D., Liao, W., Wu, G.R., Xu, Q., Zhou, D., Chen, H.,Abnormal Functional Connectivity Density in Psychogenic Non-epileptic Seizures, Epilepsy Research (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.05.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Title page Title: Abnormal Functional Connectivity Density in Psychogenic Non‐epileptic
ip t
Seizures
cr
Running title: Abnormal FCD in PNES
us
Authors: Ju‐Rong Ding1,2, Dongmei An3, Wei Liao4,5, Guo‐Rong Wu1, Qiang Xu5, Dong
an
Zhou3*, Huafu Chen1*
M
1
Key Laboratory for Neuroinformation of Ministry of Education, School of Life Science
2
te
610054, PR China.
d
and Technology, University of Electronic Science and Technology of China, Chengdu
Ac ce p
Institute of Automation and Electronic Information, Sichuan University of Science
and Engineering, Zigong 643000, PR China.
3
Department of Neurology, West China Hospital of Sichuan University, Chengdu
610041, PR China.
4
Center for Cognition and Brain Disorders and the Affiliated Hospital, Hangzhou
Normal University, Hangzhou 310015, PR China. 5
Department of Medical Imaging, Nanjing Jinling Hospital, Clinical School, Medical
College, Nanjing University, Nanjing 210002, PR China. 1
Page 1 of 37
*Corresponding authors: Huafu Chen, School of Life Science and Technology, University of Electronic Science
ip t
and Technology of China, Chengdu 610054, PR China. Fax: 86‐28‐83208238. E‐mail: [email protected].
cr
Dong Zhou, Department of Neurology, West China Hospital of Sichuan University,
us
Chengdu 610041, PR China. Fax: 86‐28‐85422549. E‐mail: [email protected].
Ac ce p
te
d
M
an
2
Page 2 of 37
Abstract Purpose: Psychogenic non‐epileptic seizures (PNES) are paroxysmal behaviors that
ip t
resemble epileptic seizures but lack abnormal electrical activity. Some neuroimaging studies have reported that PNES exhibits aberrant functional connectivity in specific
cr
brain networks. Thus, advanced neuroimaging technologies may aid clinical diagnosis
us
and treatment of PNES.
Methods: We investigated changes in brain functional connectivity in 18 patients
an
with PNES and 20 healthy controls. Functional connectivity density mapping (FCDM),
M
a voxelwise data‐driven technique, was employed to compute local and global FCD maps. Then, short‐range and long‐range FCD values were calculated and group
d
analyses performed between patents with PNES and healthy controls. A correlation
te
analysis with clinical variables was also performed.
Ac ce p
Results: We found that patients with PNES showed abnormal FCD regions mainly in the frontal cortex, sensorimotor cortex, cingulate gyrus, insula and occipital cortex.
Seed‐voxel correlation analyses also showed disrupted functional connectivity between these regions. In addition, the occipital cortex FCD correlated with duration of disease.
Conclusion: The present results support the hypothesis that patients with PNES are associated with altered attention, sensorimotor and emotion systems. Furthermore, correlations between altered regions in the occipital cortex and duration of disease may reflect an adaptation in these patients for long‐term hypervigilance and 3
Page 3 of 37
increased response to external stimuli. This study adds new knowledge to our understanding of the pathophysiological mechanisms underlying PNES.
ip t
Keywords: long‐range FCD; short‐range FCD; resting‐state functional connectivity;
cr
sensorimotor system; psychogenic non‐epileptic seizures
Ac ce p
te
d
M
an
us
4
Page 4 of 37
1.1 Introduction Psychogenic non‐epileptic seizures (PNES) are paroxysmal behaviors with altered
ip t
movement, sensation, or experience, which resemble epileptic seizures, but are not accompanied by ictal epileptiform brain discharges (Baslet, 2011; Devinsky et al.,
cr
2011). PNES can be interpreted as an experiential or behavioral response to
us
emotional, psychological, or social distress (Reuber, 2008). Patients with PNES are often misdiagnosed and treated for epilepsy, which is detrimental because of the side
an
effects of antiepileptic drugs and the delay in proper treatment(Leis et al., 1992; Reuber et al., 2004). It has been observed that the diagnosis of PNES is usually
M
delayed for an average of 7 years (Reuber et al., 2002), significantly impacting
te
d
patients’ quality of life (Szaflarski et al., 2003b; Szaflarski et al., 2003a).
Etiologically, PNES is related to the dysfunction in processing of psychological or
Ac ce p
social distress (Baslet, 2011; Uliaszek et al., 2012), manifesting as an altered cognitive‐emotional attention system (Baslet, 2011). During episodes of PNES, sensorimotor and cognitive processes are affected and not properly integrated,
resulting in a range of involuntary behavioral patterns (Baslet, 2011). Findings from
recent neuroimaging studies have provided evidence to support this notion. Using
resting‐state functional magnetic resonance imaging (rs‐fMRI) technology, van der Kruijs et al. found dysfunctional connectivity between emotional, executive control, and sensorimotor networks in PNES (van der Kruijs et al., 2012). In addition, an EEG synchronization study revealed decreased prefrontal and parietal synchronization in 5
Page 5 of 37
PNES, reflecting dysfunction of fronto‐parietal networks (Knyazeva et al., 2011). Overall, these studies suggest aberrant functional connectivity in specific brain
ip t
networks, contributing to understanding the pathophysiological mechanism of PNES. Therefore, advanced neuroimaging technologies may aid the clinical diagnosis and
cr
treatment of PNES.
us
In the present study, we compared brain functional connectivity density (FCD)
an
between patients with PNES and healthy controls. Functional connectivity density mapping (FCDM) is a voxelwise data‐driven technique, recently proposed by Tomasi
M
and Volkow (Tomasi and Volkow, 2010). Simple voxelwise functional connectivity analyses only depict brain functional connectivity. In contrast, FCDM can locate highly
te
d
connected brain regions (functional hubs), and is broadly equivalent to the combination of simple voxelwise functional connectivity analyses and graph theory
Ac ce p
analysis. In addition, simple voxelwise functional connectivity analyses are
computationally demanding. However, FCDM is an ultrafast method and can compute local and global FCD maps with high spatial resolution (3‐mm isotropic), allowing identification of functional hubs (regions that are densely connected) with high sensitivity and discrimination among short‐range FCD hubs and long‐range FCD
hubs (Tomasi and Volkow, 2012b,a). Thus, the present study used FCDM analysis to investigate abnormal connectivity in PNES. We aimed to find regions exhibiting altered FCD in patients with PNES, and further investigate whether altered brain regions were related to the duration of disease. On the basis of previous findings and 6
Page 6 of 37
clinical symptoms of PNES, we hypothesized that regions with altered FCD in patients with PNES might be associated with attention, emotion, and sensorimotor systems.
ip t
1.2 Materials and Methods
cr
1.2.1 Participants
us
The participants were the same as in our previous study (Ding et al., 2013). A total of 20 patients with PNES (7 males, mean age: 19.65±7.56 years) and 20 healthy
an
volunteers (8 males, mean age: 21.85±1.70 years) from the Department of Neurology, West China Hospital, Chengdu, China were recruited. Patients with PNES were given
M
definitive diagnoses by experienced neurologists using clinical descriptions of symptoms and long‐term video/EEG monitoring, consistent with recent
te
d
recommendations (Benbadis et al., 2004; Devinsky et al., 2011). The inclusion criteria were: 1) at least one single typical episode recorded by video EEG, and EEG did not
Ac ce p
show any epileptiform discharge or ictal slowing; 2) patients had no history of neurological disease; 3) patients had no obvious abnormality in routine structural MRI examinations. The exclusion criteria were: 1) patients with neurological
comorbidity (e.g. epilepsy); 2) patients with malingering, or any psychiatric disorders
(e.g. mood and anxiety disorders, schizophrenia, and psychosis). Here, the diagnosis
of malingering or psychiatric disorders was determined by two attending psychiatrists using the Structured Clinical Interview for DSM‐IV (SCID)‐Patients Version and their scores on the Hamilton Anxiety Rating Scale and Hamilton Depression Rating Scale. Only patients with a diagnosis of definite PNES were included in the study. Four out 7
Page 7 of 37
of 20 patients were taking antiepileptic drugs before the diagnosis of PNES. All drugs were discontinued at least 2 weeks prior to MRI examination. Demographic and
ip t
Clinical Characteristics of the patients with PNES are shown in Table 1. To increase the homogeneity of the patient group, two patients with very long duration of disease
cr
(approximately 18 years) were excluded from the study, resulting in a final analysis of
us
18 patients with PNES and 20 healthy controls. The control subjects had no neurologic/psychiatric disorders, evaluated using the SCID‐Non‐Patient Version, and
an
had not taken any psychotropic medication within the past 6 months. This study was approved by the Local Ethics Committee of West China Hospital, and written
M
informed consent was obtained from each subject before experimentation.
te
1.2.2 Data Acquisition
d
All imaging data were collected on a 3T Siemens Trio system (Erlangen, Germany) at
Ac ce p
the MR Research Center of West China Hospital, Chengdu, China. During data
acquisition, subjects were instructed to relax with their eyes closed, and to keep their heads still. Foam padding and earplugs were used to reduce head motion and scanner noise. Functional images were acquired using a single‐shot, gradient‐recalled echo planar imaging sequence for a total of 205 volumes (repetition time [TR]/echo
time [TE] = 2000/30 ms; flip angle = 90°; field of view = 240×240 mm2; in‐plane matrix = 64×64; voxel size = 3.75×3.75×5 mm3, no slice gap; 30 axial slices). Additionally, high‐resolution T1‐weighted anatomical images were also acquired using a magnetization‐prepared rapid gradient‐echo sequence (TR/TE = 20/3.69 ms; flip angle 8
Page 8 of 37
= 12°; field of view = 250×250 mm2; in‐plane matrix = 320×320; voxel size = 0.78×0.78×1 mm3, no slice gap, 128 sagittal slices) for each subject.
ip t
1.2.3 Data Preprocessing
cr
Functional images preprocessing was performed using the Statistical Parametric
us
Mapping software (SPM8, http://www.fil.ion.ucl.ac.uk/spm). The first five volumes were not analyzed to allow for signal equilibration effects. The remaining 200
an
consecutive volumes were corrected for temporal differences and head motion, spatially normalized to the Montreal Neurological Institute (MNI) echo‐planar
M
imaging template and resampled to 3‐mm cubic voxels. There were no subjects with movement greater than 1.5 mm translation or 1.5° rotation. Recent studies have
te
d
shown that functional connectivity analysis is sensitive to gross head motion effects (Power et al., 2012; Van Dijk et al., 2012); therefore, we further evaluated the mean
Ac ce p
absolute displacement of each brain volume as compared with the previous volume(Van Dijk et al., 2012). The largest mean displacement (MD) of all subjects was
less than 0.2 mm, and there was no significant difference in mean displacement between patients with PNES and healthy controls (MD: 0.0838±0.0319 for PNES and 0.0869±0.0316 for healthy controls; p = 0.7568 ) using a two‐sample two‐tailed t test. For each subject, we extracted all voxels’ time series data from brain gray matter (not including the cerebellum). In the present study, we only investigated cerebral
alterations in patients with PNES, thus the cerebellum was excluded. Next, the time series of each voxel was corrected using a linear regression process to remove several 9
Page 9 of 37
spurious sources of variances, including six head motion parameters and averaged signals from ventricles and white matter. The residuals of these regressions were
ip t
temporally band‐pass filtered (0.01‐0.08 Hz) to reduce low‐frequency drift (Foerster et al., 2005) and high‐frequency noise related to respiratory and other physiological
cr
processes (Cordes et al., 2001).
us
1.2.4 Functional Connectivity Density
an
The preprocessed image data underwent FCD mapping (Tomasi and Volkow, 2010) to compute the strength of the local FCD (lFCD) and global FCD (gFCD), a detailed
M
description of the computation of lFCD and gFCD is given in Tomasi and Volkow (2010). Briefly, we first defined the number of functional connections, ki , which was
te
d
calculated using Pearson correlation coefficients between the time series of voxel i and the other voxels using a given threshold T . Here, the correlation coefficient
Ac ce p
threshold T was set to 0.44, corresponding to p < 0.05 , family‐wise error (FWE)‐corrected. This correlation coefficient threshold was used to reduce the
chance of false positive connections across all subjects.
According to the definition of Tomasi and Volkow (2010), the lFCD at a given voxel i is the local ki between i and its neighbor voxels using a three‐dimensional searching algorithm developed in Interactive Data Language (IDL). When all the neighbors of voxel i are detected, the local ki can be calculated. Then the same steps are repeated for the next voxel. After obtaining all voxels’ local functional 10
Page 10 of 37
connections, the lFCD mapping of each subject was obtained.
ip t
For a given voxel i , the gFCD was defined as the global functional connections, ki , between i and all other voxels in the brain. Two voxels were considered
cr
functionally connected if the correlation coefficient was larger than 0.44. The
us
calculation was repeated for all voxels in the brain. A simple approach, based on a parallel algorithm, was developed in C‐language to speed up calculation of the gFCD
an
by taking advantage of multiprocessor computer architectures.
1.2.5 Short‐ and Long‐range FCD
M
The lFCD predominantly reflects functional connectivity of the local cluster, so the
te
d
short‐range FCD is equated to the lFCD. As the gFCD included both local and distal functional connections, the long‐range FCD is defined as gFCD‐lFCD (Tomasi and
Ac ce p
Volkow, 2012b,a). To minimize differences in functional anatomy of the brain across
subjects, short‐ and long‐range FCD maps were spatially smoothed by convolution
with an isotropic Gaussian kernel (FWHM=8mm) in SPM8 (Tomasi and Volkow, 2012b,a). For each subject, short‐ and long‐range FCD distributions were further scaled by the average strength in the whole brain respectively to reduce individual
overall differences in the strength of FCD. 1.2.6 Statistical analysis One‐way analysis of covariance (ANCOVA) with three covariates (age, gender and 11
Page 11 of 37
mean displacement) was implemented in SPM8 to compare group differences in short‐ and long‐range FCD respectively. The statistical significance of group differences was set at t > 2.4377 (individual voxel threshold p < 0.01 , df = (1,36) )
ip t
and minimum cluster size of 59 voxels, corresponding to a corrected p < 0.05 . This
cr
correction was confined within the brain gray matter (not including the cerebellum)
us
and was determined by the Monte Carlo simulations performed by the AlphaSim program in the REST toolkit (http://sourceforge.net/projects/resting‐fmri).
an
Further, clusters with significant differences were evaluated using regions of interest
M
(ROI) analysis. Each ROI was a 9‐mm isotropic cube including 27 voxels (0.73cm3), centered at the Montreal Neurological Institute coordinates of the local maxima. A
te
d
correlation analysis was then performed between these ROIs and duration of disease. Using standard seed‐voxel correlation analyses, we further examined resting‐state
Ac ce p
functional connectivity (RSFC) maps through Pearson correlation coefficients between the averaged time series of the abnormal ROIs and those of other voxels in the brain gray matter, respectively. The correlation coefficient maps were converted
into z maps by Fisher’s r ‐to‐ z transformation to improve normality, and the z maps were spatially smoothed (FWHM=8mm) like the FCD maps. Group comparisons were performed using one‐way ANCOVA, and the statistical threshold was set at
p < 0.05 (corrected for multiple comparisons using the AlphaSim program).
1.3 Results 12
Page 12 of 37
1.3.1 Short‐ and long‐range FCD distributions Fig. 1A and B show the average distribution of short‐range FCD in patients with PNES
ip t
and healthy controls respectively. For both patients with PNES and healthy controls, the high short‐range FCD was mainly distributed in the bilateral precuneus, cuneus,
cr
occipital cortex, parietal cortex, postcentral gyrus, and middle temporal gyrus. Fig. 2A
us
and B show the average distribution of long‐range FCD in patients with PNES and healthy controls, respectively. The high long‐range FCD was also distributed in the
an
bilateral posterior cingulate/precuneus, median cingulate gyrus, parietal cortex, angular, middle temporal gyrus and dorsolateral prefrontal cortex. Regions with high
M
FCD indicate that they might play important roles in brain networks, namely hubs. The FCD hubs found in the present study are consistent with previous studies
te
d
(Buckner et al., 2009; Tomasi and Volkow, 2012b).
Ac ce p
1.3.2 Group Comparisons
We found that patients with PNES showed increased short‐range FCD mainly in the
left middle frontal gyrus, superior frontal gyrus, medial part of the superior frontal
gyrus, anterior cingulate gyrus, supplementary motor area, and bilateral median cingulate gyrus, and decreased short‐range FCD mainly in the right middle occipital
gyrus compared with healthy controls (Fig. 1C, p < 0.05 , AlphaSim corrected). The results of ROI analysis are shown in Table 2. Compared with healthy controls, patients with PNES showed increased long‐range 13
Page 13 of 37
FCD mainly in the bilateral calcarine fissure, lingual gyrus, supplementary motor area, and the right superior temporal gyrus, insula, precentral and postcentral gyrus, and
ip t
the left paracentral lobule; and decreased long‐range FCD mainly in the right medial prefrontal cortex, middle frontal gyrus, triangular and opercular parts of the inferior
cr
frontal gyrus, superior frontal gyrus, medial part of the superior frontal gyrus,
us
supramarginal gyrus and inferior parietal gyrus (Fig. 2C, p < 0.05 , AlphaSim corrected). The results of the ROI analysis are shown in Table 3.
an
1.3.3 RSFC Networks
M
To further explore the hypo/hyper connectivity between brain regions and abnormal short‐ and long‐range FCD regions in patients with PNES, we examined functional
te
d
connectivity networks in patients and healthy controls.
Ac ce p
The RSFC maps for abnormal FCD regions belonging to the same brain system, i.e. the left frontal, cingulate, visual, sensorimotor, and right frontal cortex, were combined using a conjunction analysis (Fox et al., 2005; Zhang et al., 2011). For the RSFC
conjunctions of the left middle, superior, and medial part of the superior frontal gyrus seeds, we found increased functional connectivity among those seeds and in
the anterior cingulate gyrus and right median cingulate gyrus in patients with PNES (Fig. 3A, p < 0.05 , AlphaSim corrected). For the RSFC conjunctions of the anterior cingulate gyrus and bilateral median cingulate gyrus seeds, we observed that the patients with PNES exhibited increased functional connectivity mainly between each 14
Page 14 of 37
other and in the frontal cortex, supplementary motor area and precuneus (Fig. 3B,
p < 0.05 , AlphaSim corrected). For the RSFC conjunctions of the bilateral
ip t
supplementary motor area, right precentral and postcentral gyrus, and left paracentral lobule seeds, we found increased functional connectivity mainly between
cr
each other and in the bilateral precuneus, median cingulate gyrus, superior temporal
us
cortex, and right pole part of the superior temporal cortex (Fig. 3D, p < 0.05 , AlphaSim corrected) in the patients with PNES. For the RSFC conjunctions of the
an
bilateral calcarine fissure and lingual gyrus seeds, we found increased functional connectivity between each other and in the bilateral cuneus, postcentral gyrus and
M
left precuneus, and decreased functional connectivity in the bilateral fusiform gyrus,
te
patients with PNES.
d
middle and superior occipital cortex (Fig. 3E, p < 0.05 , AlphaSim corrected) in the
Ac ce p
For the RSFC conjunctions of the right medial prefrontal cortex, middle frontal gyrus,
triangular and opercular parts of the inferior frontal gyrus, superior and medial parts
of the superior frontal gyrus seeds, we found common lower functional connectivity in the right inferior parietal, supramarginal and angular gyrus in patients with PNES (Fig. 3F, p < 0.05 , AlphaSim corrected).
The RSFC maps of the right insula exhibited increased functional connectivity in the bilateral superior temporal cortex, supplementary motor area, rolandic area and left supramarginal gyrus (Fig. 3C, p < 0.05 , AlphaSim corrected). 15
Page 15 of 37
1.3.4 Correlation with Duration of Disease
ip t
Spearman correlations were calculated between the abnormal regions and duration of disease after removing potential outliers (Schwarzkopf et al., 2012). As seen in Fig.
cr
4, three regions with increased long‐range FCD in patients with PNES were
us
significantly positively correlated with disease duration ( p < 0.05 ), including the bilateral lingual gyrus, and the right calcarine fissure. For short‐range FCD, no regions
an
were found to correlate with disease duration.
M
1.4 Discussion
The present study investigated map changes in brain functional connectivity,
te
d
including short‐ and long‐range FCD, in patients with PNES using a voxelwise data‐driven approach. The regions with abnormal functional connectivity density in
Ac ce p
patients with PNES were mainly involved in the frontal cortex, sensorimotor cortex, cingulate gyrus, insula, and occipital cortex, supporting the hypothesis that PNES are
associated with altered attention, sensorimotor and emotion systems. Furthermore,
some altered regions in the occipital cortex were correlated with duration of disease. This study provides new insights into our understanding of the pathophysiological mechanisms underlying PNES. 1.4.1 Altered Short‐range FCD Compared with healthy controls, regions with increased short‐range FCD in patients 16
Page 16 of 37
with PNES were mainly involved in the left frontal cortex, anterior cingulate gyrus, and the bilateral median cingulate gyrus (Table 2). The frontal cortex is considered
ip t
the emotional control center, and the left frontal cortex is responsible for language skills, problem solving ability, impulse control, and judgment (Miller and Cummings,
cr
1999). The cingulate cortex is engaged in self‐control of cognitive processing and
us
attention, novelty detection, and task implementation (Bush et al., 2000; Dosenbach et al., 2006; Paus et al., 1993; Paus, 2001). Previous studies have found that patients
an
with PNES show hyper responses to novel stimuli, and fail to neglect irrelevant stimuli with social emotion (Gene‐Cos et al., 2005; Pouretemad et al., 1998).
M
Moreover, patients with PNES experience more stress, are more likely to deny their experiences of psychological stress, and to choose avoidance oriented coping
te
d
strategies, such as behavioral efforts to avoid conflicts or stress (Bakvis et al., 2011; Frances et al., 1999; Tojek et al., 2000). The findings of increased short‐range FCD in
Ac ce p
the left frontal and cingulate cortex in patients with PNES suggest that the function of
these regions may be impaired. Furthermore, the RSFC maps based on seeds in the left frontal and cingulate cortex exhibited hyperconnectivity within each system and
also between each other. Therefore, our findings support an altered
cognitive‐emotional attention system in patients the PNES (Baslet, 2011).
1.4.2 Altered Long‐range FCD Patients with PNES showed increased long‐range FCD mainly in the insula, sensorimotor (including bilateral supplementary motor area, right precentral and 17
Page 17 of 37
postcentral gyrus) and occipital cortex (including bilateral calcarine fissure and lingual gyrus) (Table 3). The insula is an important multisensory integration area that
ip t
mediates interpretation of sensory information from the body, related to emotion regulation, visceral sensory perception, and self‐awareness (Craig, 2002; Pollatos et
cr
al., 2007). Altered long‐range FCD found in the insula supports that PNES is involved
us
in impaired function of emotion and cognition (Reuber et al., 2004).
an
PNES are episodes of altered movement and sensation (Brown et al., 2011). During episodes of PNES, sensorimotor and cognitive processes are affected and not
M
properly integrated, resulting in a range of involuntary behavioral patterns (Baslet, 2011). For most patients, these involuntary behavioral patterns are avoidant coping
te
d
strategies used to deal with stressors, to keep them from experiencing stressful events or from memories of those events (Bakvis et al., 2011; Magaudda et al., 2011).
Ac ce p
A recent rs‐fMRI study reported that patients with PNES showed abnormal functional connectivity in the sensorimotor cortex (van der Kruijs et al., 2012). In addition, Labate et al. (2012) found abnormal cortical atrophy of motor regions. Combining
these previous findings and the known symptoms of PNES, our results of altered
long‐range FCD in the sensorimotor cortex provide further evidence that the sensorimotor cortex plays an important role in the pathogenesis of PNES. Moreover, seed‐voxel correlation analysis showed hyperconnectivity between the insula and supplementary motor area, and between the sensorimotor and cingulate cortex, supporting dysfunctional connectivity between emotional and sensorimotor 18
Page 18 of 37
networks in PNES (van der Kruijs et al., 2012). The present findings also provide neuroimaging evidence for understanding the pathology of PNES, as PNES are
in the processing of psychological or social distress (Baslet, 2011).
cr
ip t
associated with involuntary changes in movement and sensation due to a dysfunction
us
The calcarine fissure and lingual gyrus are thought to be related to visual memory and face memory (Kapur et al., 1995). In an experimental study, patients with PNES
an
were found to show high attentional bias to angry faces, and were likely to take avoidant strategies, i.e. related motor responses, to deal with the angry faces (Bakvis
M
et al., 2011). In the present study, we found that patients with PNES exhibited altered long‐range FCD in the occipital cortex, and the alterations correlated with duration of
te
d
disease, which might reflect an adaptation in patients with PNES for long‐term hypervigilance and increased response to external stimuli.
Ac ce p
In addition, we found regions with decreased long‐range FCD mainly in the right frontal cortex in patients with PNES. The frontal cortex plays an important role in maintaining long‐term memories, which are involved in emotions derived from the
limbic system (Buckner and Petersen, 1996). The right frontal cortex is associated with social cognition and emotions, especially for negative emotions (Miller and Cummings, 1999). A previous study demonstrated that cognitive integrative functions for dealing with social stress and memory are impaired in patients with PNES (Bakvis
et al., 2010). Our findings further support this finding. For the seed‐voxel correlation 19
Page 19 of 37
analyses, patients with PNES showed decreased functional connectivity between the right frontal and parietal cortex (including the right inferior parietal, supramarginal,
ip t
and angular gyrus). The fronto‐parietal network is associated with attentional control, and is activated when attention is directed to external stimuli in cognitive tasks (Fox
cr
et al., 2005; Seeley et al., 2007; Sridharan et al., 2008). Therefore, decreased
us
functional connectivity between the right frontal and parietal cortex indicates that the fronto‐parietal network may be impaired, providing evidence for dysfunction of
an
cognitive attention and executive control in patients with PNES (Baslet, 2011).
M
Some methodological limitations should be mentioned in this study. First, as our sample size was modest, a relatively weak correction strategy (AlphaSim program)
te
d
was used for multiple comparisons in this study. Thus, the present results require replication using larger sample sizes. Second, recent studies have shown significant
Ac ce p
effects of head motion on functional connectivity analysis (Power et al., 2012; Van Dijk et al., 2012). We examined the mean displacement of head motion and found no significant between‐group difference. Nevertheless, we partialed out the effect of
mean displacement of head motion in our statistical comparisons to reduce the effect of head motion. Third, the correlations between disease duration and regions
with altered FCD were not corrected for multiple comparisons. Finally, as this was an exploratory study, we only selected a single threshold to calculate FCD maps. In addition, a range of thresholds can also be used to test the stability of the results. 20
Page 20 of 37
1.5 Conclusion In conclusion, we investigated the map changes in brain functional connectivity in
ip t
patients with PNES. The abnormal FCD regions and functional connectivity based on these regions in patients with PNES were mainly associated with attention,
cr
sensorimotor, and emotion systems, supporting our hypothesis. Furthermore, the
us
altered regions were also involved in the occipital cortex, and some alterations correlated with duration of disease, which might reflect an adaptation in patients
an
with PNES for long‐term hypervigilance and increased response to external stimuli. This study improves our understanding of the pathophysiological mechanisms
Ac ce p
te
d
M
underlying PNES.
21
Page 21 of 37
Acknowledgments: This work was supported by the 973 Project (2012CB517901); the Natural Science Foundation of China (61035006, 61125304); and the Academic
Ac ce p
te
d
M
an
us
cr
Author Disclosure Statement: No competing financial interests exist.
ip t
new artist Ministry of Education doctoral post graduate grant (A03003023901004).
22
Page 22 of 37
References
Ac ce p
te
d
M
an
us
cr
ip t
Bakvis, P., Spinhoven, P., Putman, P., Zitman, F. G., Roelofs, K., 2010. The effect of stress induction on working memory in patients with psychogenic nonepileptic seizures. Epilepsy Behav 19, 448-454. Bakvis, P., Spinhoven, P., Zitman, F. G., Roelofs, K., 2011. Automatic avoidance tendencies in patients with psychogenic non-epileptic seizures. Seizure 20, 628-634. Baslet, G., 2011. Psychogenic non-epileptic seizures: a model of their pathogenic mechanism. Seizure 20, 1-13. Benbadis, S. R., Siegrist, K., Tatum, W. O., Heriaud, L., Anthony, K., 2004. Short-term outpatient EEG video with induction in the diagnosis of psychogenic seizures. Neurology 63, 1728-1730. Brown, R. J., Syed, T. U., Benbadis, S., LaFrance, W. C., Jr., Reuber, M., 2011. Psychogenic nonepileptic seizures. Epilepsy Behav 22, 85-93. Buckner, R. L., Petersen, S. E., 1996. What does neuroimaging tell us about the role of prefrontal cortex in memory retrieval? Seminars in the Neurosci 8, 47-55. Buckner, R. L., Sepulcre, J., Talukdar, T., Krienen, F. M., Liu, H., Hedden, T., Andrews-Hanna, J. R., Sperling, R. A., Johnson, K. A., 2009. Cortical hubs revealed by intrinsic functional connectivity: mapping, assessment of stability, and relation to Alzheimer's disease. J Neurosci 29, 1860-1873. Bush, G., Luu, P., Posner, M. I., 2000. Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn Sci 4, 215-222. Cordes, D., Haughton, V. M., Arfanakis, K., Carew, J. D., Turski, P. A., Moritz, C. H., Quigley, M. A., Meyerand, M. E., 2001. Frequencies contributing to functional connectivity in the cerebral cortex in "resting-state" data. AJNR Am J Neuroradiol 22, 1326-1333. Craig, A. D., 2002. How do you feel? Interoception: the sense of the physiological condition of the body. Nat Rev Neurosci 3, 655-666. Devinsky, O., Gazzola, D., LaFrance, W. C., Jr., 2011. Differentiating between nonepileptic and epileptic seizures. Nat Rev Neurol 7, 210-220. Ding, J. R., An, D., Liao, W., Li, J., Wu, G. R., Xu, Q., Long, Z., Gong, Q., Zhou, D., Sporns, O., Chen, H., 2013. Altered functional and structural connectivity networks in psychogenic non-epileptic seizures. PLoS One 8, e63850. Dosenbach, N. U., Visscher, K. M., Palmer, E. D., Miezin, F. M., Wenger, K. K., Kang, H. C., Burgund, E. D., Grimes, A. L., Schlaggar, B. L., Petersen, S. E., 2006. A core system for the implementation of task sets. Neuron 50, 799-812. Foerster, B. U., Tomasi, D., Caparelli, E. C., 2005. Magnetic field shift due to mechanical vibration in functional magnetic resonance imaging. Magn Reson Med 54, 1261-1267. Fox, M. D., Snyder, A. Z., Vincent, J. L., Corbetta, M., Van Essen, D. C., Raichle, M. E., 2005. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci U S A 102, 9673-9678. Frances, P. L., Baker, G. A., Appleton, P. L., 1999. Stress and avoidance in Pseudoseizures: testing the assumptions. Epilepsy Res 34, 241-249. Gene-Cos, N., Pottinger, R., Barrett, G., Trimble, M. R., Ring, H. A., 2005. A comparative study of mismatch negativity (MMN) in epilepsy and non-epileptic seizures. Epileptic Disord 7, 363-372. Kapur, N., Friston, K. J., Young, A., Frith, C. D., Frackowiak, R. S., 1995. Activation of human hippocampal formation during memory for faces: a PET study. Cortex 31, 99-108. Knyazeva, M. G., Jalili, M., Frackowiak, R. S., Rossetti, A. O., 2011. Psychogenic seizures and frontal disconnection: EEG synchronisation study. J Neurol Neurosurg Psychiatry 82, 505-511. Labate, A., Cerasa, A., Mula, M., Mumoli, L., Gioia, M. C., Aguglia, U., Quattrone, A., Gambardella, A., 2012. Neuroanatomic correlates of psychogenic nonepileptic seizures: A cortical thickness and VBM study. Epilepsia 53, 377-385. Leis, A. A., Ross, M. A., Summers, A. K., 1992. Psychogenic seizures: ictal 23
Page 23 of 37
Ac ce p
te
d
M
an
us
cr
ip t
characteristics and diagnostic pitfalls. Neurology 42, 95-99. Magaudda, A., Gugliotta, S. C., Tallarico, R., Buccheri, T., Alfa, R., Lagana, A., 2011. Identification of three distinct groups of patients with both epilepsy and psychogenic nonepileptic seizures. Epilepsy Behav 22, 318-323. Miller, B. L., Cummings, J. L., 1999. The human frontal lobes : functions and disorders. Guilford Press: New York. Paus, T., Petrides, M., Evans, A. C., Meyer, E., 1993. Role of the human anterior cingulate cortex in the control of oculomotor, manual, and speech responses: a positron emission tomography study. J Neurophysiol 70, 453-469. Paus, T., 2001. Primate anterior cingulate cortex: where motor control, drive and cognition interface. Nat Rev Neurosci 2, 417-424. Pollatos, O., Gramann, K., Schandry, R., 2007. Neural systems connecting interoceptive awareness and feelings. Hum Brain Mapp 28, 9-18. Pouretemad, H. R., Thompson, P. J., Fenwick, P. B., 1998. Impaired sensorimotor gating in patients with non-epileptic seizures. Epilepsy Res 31, 1-12. Power, J. D., Barnes, K. A., Snyder, A. Z., Schlaggar, B. L., Petersen, S. E., 2012. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. Neuroimage 59, 2142-2154. Reuber, M., Fernandez, G., Bauer, J., Helmstaedter, C., Elger, C. E., 2002. Diagnostic delay in psychogenic nonepileptic seizures. Neurology 58, 493-495. Reuber, M., Baker, G. A., Gill, R., Smith, D. F., Chadwick, D. W., 2004. Failure to recognize psychogenic nonepileptic seizures may cause death. Neurology 62, 834-835. Reuber, M., 2008. Psychogenic nonepileptic seizures: answers and questions. Epilepsy Behav 12, 622-635. Schwarzkopf, D. S., De Haas, B., Rees, G., 2012. Better ways to improve standards in brain-behavior correlation analysis. Front Hum Neurosci 6, 200. Seeley, W. W., Menon, V., Schatzberg, A. F., Keller, J., Glover, G. H., Kenna, H., Reiss, A. L., Greicius, M. D., 2007. Dissociable intrinsic connectivity networks for salience processing and executive control. J Neurosci 27, 2349-2356. Sridharan, D., Levitin, D. J., Menon, V., 2008. A critical role for the right fronto-insular cortex in switching between central-executive and default-mode networks. Proc Natl Acad Sci U S A 105, 12569-12574. Szaflarski, J. P., Hughes, C., Szaflarski, M., Ficker, D. M., Cahill, W. T., Li, M., Privitera, M. D., 2003a. Quality of life in psychogenic nonepileptic seizures. Epilepsia 44, 236-242. Szaflarski, J. P., Szaflarski, M., Hughes, C., Ficker, D. M., Cahill, W. T., Privitera, M. D., 2003b. Psychopathology and quality of life: psychogenic non-epileptic seizures versus epilepsy. Med Sci Monit 9, CR113-118. Tojek, T. M., Lumley, M., Barkley, G., Mahr, G., Thomas, A., 2000. Stress and other psychosocial characteristics of patients with psychogenic nonepileptic seizures. Psychosomatics 41, 221-226. Tomasi, D., Volkow, N. D., 2010. Functional connectivity density mapping. Proc Natl Acad Sci U S A 107, 9885-9890. Tomasi, D., Volkow, N. D., 2012a. Aging and functional brain networks. Mol Psychiatry 17, 471, 549-458. Tomasi, D., Volkow, N. D., 2012b. Abnormal functional connectivity in children with attention-deficit/hyperactivity disorder. Biol Psychiatry 71, 443-450. Uliaszek, A. A., Prensky, E., Baslet, G., 2012. Emotion regulation profiles in psychogenic non-epileptic seizures. Epilepsy Behav 23, 364-369. van der Kruijs, S. J., Bodde, N. M., Vaessen, M. J., Lazeron, R. H., Vonck, K., Boon, P., Hofman, P. A., Backes, W. H., Aldenkamp, A. P., Jansen, J. F., 2012. Functional connectivity of dissociation in patients with psychogenic non-epileptic seizures. J Neurol Neurosurg Psychiatry 83, 239-247. Van Dijk, K. R., Sabuncu, M. R., Buckner, R. L., 2012. The influence of head motion on intrinsic functional connectivity MRI. Neuroimage 59, 431-438. Zhang, Z., Liao, W., Zuo, X. N., Wang, Z., Yuan, C., Jiao, Q., Chen, H., Biswal, B. B., Lu, G., Liu, Y., 2011. Resting-state brain organization revealed by functional 24
Page 24 of 37
covariance networks. PLoS One 6, e28817.
Ac ce p
te
d
M
an
us
cr
ip t
25
Page 25 of 37
Table 1 Demographic and Clinical Characteristics of patients with PNES Previous Patient
Age
Gender
Duration
Type of symptoms
1
34y
F
18y
Unresponsiveness/eye closure
2
17y
F
2y
Unresponsiveness/eye
ip t
Treatment* None
cr
None
rigidity
us
closure/hyperventilation/body
38y
F
18y
Unresponsiveness/hypermotor EX
4
23y
F
8y
Unresponsiveness/eye
an
3
None None
M
closure/hyperventilation/body
F
6
20y
M
7
14y
F
8
17y
F
1m
Unresponsiveness/eye closure
None
2y
Unresponsiveness
VPA
2m
Unresponsiveness
None
2m
Unresponsiveness/eye
None
te
13y
Ac ce p
5
d
rigidity
closure/body rigidity/trembling EX
9
14y
M
5m
Unresponsiveness/vocalization
VPA
10
16y
F
2y
Unresponsiveness/hyperventilation
None
/hypermotor EX 11
17y
F
4m
Unresponsiveness/hyperventilation
None
12
21y
F
8m
Unresponsiveness/hypermotor EX
None
13
21y
M
1m
Eye closure/hyperventilation/body None 26
Page 26 of 37
rigidity 14
13y
M
1y
Unresponsiveness/eye
None
15
13y
F
1y
ip t
closure/hyperventilation Unresponsiveness/eye
None
35y
M
15d
Unresponsiveness/eye
us
16
cr
closure/hyperventilation
None
closure/hypermotor EX F
3y
Unresponsiveness
18
20y
M
2y
Unresponsiveness/hypermotor EX
VPA
19
13y
M
1m
Unresponsiveness
None
20
18y
F
7m
Unresponsiveness/eye closure
None
CBZ
d
M
an
16y
te
17
*: All drugs were discontinued at least 2 weeks before MRI examination.
Ac ce p
Abbreviations: F: female; M: male; d: day; m: month; y: year; Hypermotor EX: Hypermotor movements of the extremities; Trembling EX: Trembling of the
extremities; VPA: valproate; CBZ: carbamazepine.
27
Page 27 of 37
Table 2 Statistical Significance (t Score; one‐way ANOVA) of Short‐Range FCD Differences between patients with PNES and HCs BA
x(mm)
y(mm)
z(mm)
Short‐range FCD increased regions
PNES>HCs(t)
ip t
Anatomical region
46
‐27
51
24
SFG_L
32
‐12
27
39
3.04
SFG_Med_L
32
‐3
21
42
ACG_L
32
‐9
33
27
4.01
SMA_L
6
‐3
MCG_R
—
3
MCG_L
—
us
18
45
3.09
‐24
48
4.45
‐24
48
3.89
‐81
27
‐4.35
M
an
3.58
d
0
3.82
cr
MFG_L
MOG_R
19
42
Ac ce p
te
Short‐range FCD decreased regions
Statistical values averaged in 9‐mm isotropic (cube; 27 voxels) regions of interest centered at the Montreal Neurological Institute coordinates (x, y, z) of the local maxima.
FCD, functional connectivity density; ANOVA, analysis of variance; PNES, psychogenic
non‐epileptic seizures; HCs, healthy controls; BA, Brodmann’s area; L, Left; R, Right; MFG, middle frontal gyrus; SFG, superior frontal gyrus; SFG_Med, medial part of superior frontal gyrus; ACG, anterior cingulate gyrus; SMA, supplementary motor area; MCG, median cingulate gyrus; MOG, middle occipital gyrus. 28
Page 28 of 37
Table 3 Statistical Significance (t Score; one‐way ANOVA) of Long‐Range FCD Differences between patients with PNES and HCs BA
x(mm)
y(mm)
z(mm)
Long‐range FCD increased regions
PNES>HCs(t)
ip t
Anatomical region
17
0
‐84
0
CAL_R
17
6
‐75
6
3.21
LING_R
17
6
‐69
6
4.02
LING_L
17
0
‐72
6
STG_R
22
48
INS_R
48
45
PCL_L
4
‐6
PreCG_R
6
SMA_R
us
‐9
‐6
3.46
‐9
0
3.63
‐28
75
3.07
‐21
75
3.72
d
M
an
3.26
4
24
‐33
72
2.90
—
‐3
‐12
66
3.02
6
6
‐15
78
3.11
Ac ce p
SMA_L
te
PoCG_R
21
3.99
cr
CAL_L
Long‐range FCD decreased regions MFG_Orb_R
46
39
57
‐6
‐3.92
MFG_R
10
36
57
0
‐3.37
IFG_Tri_R
48
51
21
21
‐3.10
IFG_Oper_R
44
51
21
33
‐3.36
SMG_R
40
57
‐33
42
‐3.45
IPG_R
40
39
‐45
39
‐3.38
29
Page 29 of 37
SFG_R
8
24
15
66
‐3.58
SFG_Med_R
8
12
30
63
‐3.21
ip t
Statistical values averaged in 9‐mm isotropic (cube; 27 voxels) regions of interest
cr
centered at the Montreal Neurological Institute coordinates (x, y, z) of the local
us
maxima.
FCD, functional connectivity density; ANOVA, analysis of variance; PNES, psychogenic
an
non‐epileptic seizures; HCs, healthy controls; BA, Brodmann’s area; L, Left; R, Right; CAL, calcarine fissure; LING, lingual gyrus; STG, superior temporal gyrus; INS, insula;
M
PCL, paracentral lobule; PreCG, precentral gyrus; PoCG, postcentral gyrus; SMA,
d
supplementary motor area; MFG_Orb, medial prefrontal cortex; MFG, middle frontal
te
gyrus; IFG_Tri, triangular part of inferior frontal gyrus; IFG_Oper, opercular part of inferior frontal gyrus; SMG, supramarginal gyrus; IPG, inferior parietal gyrus; SFG,
Ac ce p
superior frontal gyrus; SFG_Med, medial part of superior frontal gyrus.
30
Page 30 of 37
Captions to Figure Fig. 1 Distribution of short‐range FCD in the brain for 18 patients with PNES and 20
ip t
HCs and statistical differences ( t scores) between the groups. One‐way analysis of covariance with three covariates (age, gender, and mean motion) was used to
cr
contrast short‐range FCD maps across groups. The statistical significance of group
us
differences was set at t > 2.4377 (individual voxel threshold p < 0.01 , df = (1,36) ) and minimum cluster size of 59 voxels, corresponding to an AlphaSim corrected
an
p < 0.05 . The left hemisphere is on the left. df, degree of freedom; PNES, psychogenic non‐epileptic seizures; HCs, healthy controls.
M
Fig. 2 Distribution of long‐range FCD in the brain for 18 patients with PNES and 20
te
d
HCs and the statistical differences ( t scores) between the two groups. One‐way analysis of covariance with three covariates (age, gender, and mean motion) was
Ac ce p
used to contrast long‐range FCD maps across groups. The statistical significance of group
differences
was
set
at
t > 2.4377
(individual voxel threshold
p < 0.01 , df = (1,36) ) and minimum cluster size of 59 voxels, corresponding to an
AlphaSim corrected p < 0.05 . The left hemisphere is on the left. df, degree of freedom; PNES, psychogenic non‐epileptic seizures; HCs, healthy controls.
Fig. 3 Statistical significance of RSFC maps for the abnormal FCD regions between patients with PNES and HCs ( p < 0.05 , AlphaSim corrected). Group differences in conjunction analyses of the RSFC maps with seeds belonging to the left frontal cortex 31
Page 31 of 37
(A), cingulate cortex (B), right insula (C), sensorimotor cortex (D), visual cortex (E) and right frontal cortex (F), respectively. Each result is displayed on three ‘brains’ shown
ip t
from the right side (top left image of each group), from behind (top right image of each group), and from above (bottom left image of each group). Warm and cool
cr
colors indicate RSFC increases and decreases, respectively, in the patients with PNES.
us
One‐way analysis of covariance with three covariates (age, gender, and mean motion) was used to compare group differences. RSFC, resting‐state functional connectivity;
an
PNES, psychogenic non‐epileptic seizures; HCs, healthy controls.
M
Fig. 4 Correlations between regions with altered long‐range FCD and duration of disease in patients with PNES ( p < 0.05 ). Spearman correlations were calculated over
te
d
the data after removing outliers marked by circles. Cubic ROI volume: 27 voxels; ROI center coordinates shown in Table 3. CAL, calcarine fissure; LING, lingual gyrus; L, Left;
Ac ce p
R, Right.
32
Page 32 of 37
Highlights We examine changes of brain functional connectivity in PNES patients. Abnormal regions in PNES are related to attention, sensorimotor and emotion systems.
ip t
Altered regions in the occipital cortex are correlated with duration of disease.
Ac ce p
te
d
M
an
us
cr
33
Page 33 of 37
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 34 of 37
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 35 of 37
Ac ce p
te
d
M
an
us
cr
ip t
Figure(s)
Page 36 of 37
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 37 of 37
