Altered EEG resting-state large-scale brain network dynamics in euthymic bipolar disorder patients

Background Neuroimaging studies provided evidence for disrupted resting-state functional brain network activity in bipolar disorder (BD). Electroencephalographic (EEG) studies found altered temporal characteristics of functional EEG microstates during depressive episode within different affective disorders. Here we investigated whether euthymic patients with BD show deviant resting-state large-scale brain network dynamics as reflected by altered temporal characteristics of EEG microstates. Methods We used high-density EEG to explore between-group differences in duration, coverage and occurrence of the resting-state functional EEG microstates in 17 euthymic adults with BD in on-medication state and 17 age- and gender-matched healthy controls. Two types of anxiety, state and trait, were assessed separately with scores ranging from 20 to 80. Results Microstate analysis revealed five microstates (A-E) in global clustering across all subjects. In patients compared to controls, we found increased occurrence and coverage of microstate A that did not significantly correlate with anxiety scores. Conclusion Our results provide neurophysiological evidence for altered large-scale brain network dynamics in BD patients and suggest the increased presence of A microstate to be an electrophysiological trait characteristic of BD.


1
Introduction 36 Bipolar disorder (BD) is a common and severe psychiatric disorder, with an important personal and 37 societal burden (Cloutier et al., 2018;Eaton et al., 2012). The worldwide prevalence of BD is 38 considered to range between 1% and 3% (Merikangas et al., 2011;Ferrari et al., 2016). BD patients 39 are frequently misdiagnosed and often identified at late stages of disease progression, which can lead 40 to inadequate treatment (Hirschfeld, 2007) and worse functional prognosis (Vieta et al., 2018). A better 41 This is a provisional file, not the final typeset article understanding of the underlying pathophysiology is needed to identify objective biomarkers of BD that 42 would improve diagnostic and/or treatment stratification of patients. 43 Possible candidates for neurobiological biomarkers in BD could arise from the abnormalities of 44 functional brain networks. Evidence from brain imaging studies consistently points to abnormalities in 45 circuits implicated in emotion regulation and reactivity. Particularly, attenuated frontal and enhanced 46 limbic activations are reported in BD patients (Chen et al., 2011;Houenou et al., 2011;Kupferschmidt 47 and Zakzanis, 2011). Interestingly, regions implicated in the pathophysiology of the disease, such as 48 the inferior frontal gyrus, the medial prefrontal cortex (mPFC), and the amygdala present altered 49 activation patterns even in unaffected first-degree relatives of BD patients (Piguet et al., 2015), pointing 50 towards brain alterations that could underlie disease vulnerability. Moreover, evidence from functional 51 magnetic resonance imaging (fMRI) studies showed aberrant resting-state functional connectivity 52 between frontal and meso-limbic areas in BD when compared to healthy controls (Vargas et al., 2013). 53 A recently developed functional neuroanatomic model of BD suggests, more specifically, decreased 54 connectivity between ventral prefrontal networks and limbic brain regions including the amygdala 55 (Strakowski et al., 2012;Chase and Phillips, 2016). The functional connectivity abnormalities in BD 56 in brain areas associated with emotion processing were shown to vary with mood state. A resting-state 57 functional connectivity study of emotion regulation networks demonstrated that subgenual anterior 58 cingulate cortex (sgACC)-amygdala coupling is critically affected during mood episodes, and that 59 functional connectivity of sgACC plays a pivotal role in mood normalization through its interactions 60 with the ventrolateral PFC and posterior cingulate cortex (Rey et al., 2016 The patients were recruited from the Mood Disorders Unit at the Geneva University Hospital. A 99 snowball convenience sampling was used for the selection of the BD patients. Control subjects were 100 recruited by general advertisement. All subjects were clinically evaluated using clinical structured 101 interview (DIGS: Diagnostic for Genetic Studies, (Nurnberger et al., 1994). BD was confirmed in the 102 experimental group by the usual assessment of the specialized program, an interview with a 103 psychiatrist, and a semi-structured interview and relevant questionnaires with a psychologist. 104 Exclusion criteria for all participants were a history of head injury, current alcohol or drug abuse. 105 Additionally, a history of psychiatric or neurological illness and of any neurological comorbidity were 106 exclusion criteria for controls and bipolar patients, respectively. Symptoms of mania and depression 107 were evaluated using the Young Mania Rating Scale (YMRS) (Young et al., 1978) and the 108 Montgomery-Åsberg Depression Rating Scale (MADRS) (Williams and Kobak, 2008), respectively. 109 Participants were considered euthymic if they scored < 6 on YMRS and < 12 on MADRS at the time 110 of the experiment, and were stable for at least 4 weeks before. All patients were medicated, receiving 111 pharmacological therapy including antipsychotics, antidepressants and mood stabilizers, and had to be 112 under stable medication for at least 4 weeks. The experimental group included both BD I (n = 10) and 113 BD II (n = 7) types. 114 To check for possible demographic or clinical differences between groups, subject characteristics such 115 as age, education or level of depression were compared between groups using independent t-tests. 116 Anxiety is highly associated with BD (Simon et al., 2004;2007) and is a potential confounding variable 117 when investigating microstate dynamics at rest. For example, decreased duration of EEG microstates 118 at rest in patients with panic disorder has been reported (Wiedemann et al., 1998). To check for possible 119 differences in anxiety symptoms, all subjects were assessed with the State-Trait Anxiety Inventory 120 (STAI) (Spielberger et al., 1970). Anxiety as an emotional state (state-anxiety) and anxiety as a 121 personal characteristic (trait-anxiety) were evaluated separately. Scores of both state-and trait-anxiety 122 range from 20 to 80, higher values indicating greater anxiety. The scores were compared between 123 patients and controls using independent t-tests. 124 This study was carried out in accordance with the recommendations of the Ethics Committee for 125 Human Research of the Geneva University Hospital, with written informed consent from all subjects. 126 All subjects gave written informed consent in accordance with the Declaration of Helsinki. The 127 protocol was approved by the Ethics Committee for Human Research of the Geneva University 128 Hospital, Switzerland. 129

EEG recording and pre-processing 130
The EEG was recorded with a high density 256-channel system (EGI System 200; Electrical Geodesic 131 Inc., OR, USA), sampling rate of 1kHz, and Cz as acquisition reference. Subjects were sitting in a 132 This is a provisional file, not the final typeset article comfortable upright position and were instructed to stay as calm as possible, to keep their eyes closed 133 and to relax for 6 minutes. They were asked to stay awake. 134 To remove muscular artifacts originating in the neck and face the data were reduced to 204 channels. 135 Two to four minutes of EEG data were selected based on visual assessment of the artifacts and band-136 pass filtered between 1 and 40 Hz. Subsequently, in order to remove ballistocardiogram and oculo-137 motor artifacts, infomax-based Independent Component Analysis (Jung et al., 2000) was applied on all 138 but one or two channels rejected due to abundant artifacts. Only components related to physiological 139 noise, such as ballistocardiogram, saccadic eye movements, and eye blinking, were removed based on 140 the waveform, topography and time course of the component. The cleaned EEG recordings were down-141 sampled to 125 Hz and the previously identified noisy channels were interpolated using a three-142 dimensional spherical spline (Perrin et al., 1989), and re-referenced to the average reference. All the 143 preprocessing steps were done using MATLAB and the freely available Cartool Software 3.70 144 (https://sites.google.com/site/cartoolcommunity/home), programmed by Denis Brunet. 145

EEG data analysis 146
To estimate the optimal set of topographies explaining the EEG signal, a standard microstate analysis 147 was performed using k-means clustering (see Supplementary  The duration in milliseconds indicates the most common amount of time that a given microstate class 177 is continuously present. The global explained variance for a specific microstate class was calculated 178 by summing the squared spatial correlations between the representative map and its corresponding 179 assigned scalp potential maps at each time point weighted by the GFP (Murray et al., 2008

Clinical and demographic variables 199
There were no significant differences in age and level of education between the patient and the control 200 groups. In both groups, very low mean scores on depression and mania symptoms were observed, 201 which did not significantly differ between the two groups. BD patients showed higher scores on state 202 and trait scales of the STAI. For all subject characteristics, see Table 1. Since some microstate parameters showed a non-homogeneity of variances in the two groups (Levene's 214 tests for the microstate C coverage and microstates A and C duration; p<0.01), we decided to calculate 215 Mann-Whitney U test to investigate group differences for temporal parameters of each microstate. 216 We found significant between-group differences for microstate classes A and B. Both microstates 217 showed increased presence in patients in terms of occurrence and coverage. The two groups did not 218 differ in any temporal parameter of microstates C, D, or E. The results of the temporal characteristics 219 of each microstate are summarized in Table 2 and Fig. 2. 220

Clinical correlations 221
This is a provisional file, not the final typeset article The results of Spearman's rank correlation revealed no significant associations between the MADRS 230 and YMRS scores and the occurrence or coverage of the microstate A and B (all absolute r-values < 231 0.30). 232

Alpha rhythm 233
The Mann-Whitney U test showed significantly decreased alpha power (p < 0.03, Z-value 2.7) in the 234 BD compared to HC group (see Fig. 3). The results of Spearman's rank correlation revealed no 235 significant associations between the alpha power and occurrence or coverage of microstates A and B 236 (all absolute r-values < 0.40). 237 238 4 Discussion 239 Our study presents the first evidence for altered resting-state EEG microstate dynamics in euthymic 240 patients with BD. Patients were stable and did not significantly differ in their depressive or manic 241 symptomatology from healthy controls at the time of experiment. Despite this fact, they showed 242 abnormally increased presence of microstates A and B, the latter correlating with the anxiety level. 243 In an earlier combined fMRI-EEG study the microstate A was associated with the auditory network 244 (Britz et  in euthymic BD patients might be related to the hyperconnectivity of the underlying networks that 265 involve the temporal lobe, insula, mPFC, and occipital gyri. 266 Anxiety symptoms were previously associated with greater severity and impairment in BD (Simon et 267 al., 2004)  BD patients, we found an abnormally increased presence of microstate B that was associated with a 288 higher anxiety. In particular, the occurence together with coverage and only the coverage were 289 positively correlated with scores of trait and state anxiety, respectively. The observed change in 290 microstate B dynamics might be, therefore, more related to a relatively stable disposition than to the 291 actual emotional state. Previous studies also suggest that anxiety may influence visual processing 292 (Phelps et al., 2006;Laretzaki et al., 2010) and that connections between amygdala and visual cortex 293 might underlie enhanced visual processing of emotionally salient stimuli in patients with social fobia 294 (Goldin et al., 2009). Our finding of increased presence of microstate B positively associated with 295 anxiety level in euthymic BD patients is consistent with these observations. Additionally, a more 296 regular appearance for microstate B and increased overall temporal dependencies among microstates 297 were recently reported in mood and anxiety disorders, suggesting a decreased dynamicity in switching 298 between different brain states in these psychiatric conditions (Al Zoubi et al., 2019). Another 299 microstate study on anxiety disorders reported a decreased overall resting-state microstate duration in 300 panic disorder (Wiedemann et al., 1998). That early study, however, did not assess temporal 301 characteristics of different microstates separately and it is therefore difficult to compare those findings 302 with our observations. Further evidence is needed to determine, whether the increased presence of 303 microstate B in our experimental group is a characteristic feature of BD or anxiety, or whether it is 304 related to both conditions. 305 We found an unchanged duration but a higher occurence and coverage of A and B microstates in BD 306 patients. In other words, an unchanged sustainability in time and still increased presence of these 307 microstates in patients compared to healthy controls were observed. Possible explanation for this 308 finding appears to be a redundance in activation of the sensory and autobiographic memory networks were also observed in patients with multiple sclerosis, moreover predicting depression scores and other 316 clinical variables (Gschwind, et al., 2016). It was suggested that multiple sclerosis affects the "sensory" 317 (visual, auditory) rather than the higher-order (salience, central executive) functional networks (Michel 318 and Koenig, 2018 BD patients were previously shown to display lower alpha power as compared to healthy controls 335 (Basar et al., 2012), as it was the case here. We failed, however, to find any significant correlation 336 between the altered microstate dynamics and decreased alpha power. Our findings, therefore, further 337 support the previously reported independence of microstates from EEG frequency power fluctuations 338 . 339 In summary, results of the current study seem to indicate that dysfunctional activity of resting-state 340 brain networks underlying microstates A and B is a detectable impairment in BD during an euthymic 341 state. The presence of microstate A and B represents measures that might be implicated in clinical 342 practice, although using these parameters for early identification of BD at individual level could prove 343 challenging. If future studies confirm the same pattern in prodromal or vulnerable subjects, it could 344 help detection of at-risk subjects and therefore the possiblility for early intervention. The present study 345 has, however, some limitations. Our low sample size made it impossible to examine any potential 346 influence of medication on the microstate parameters by comparing patients receiving a specific drug 347 with those not receiving it. Possible effects of medication on our results should be therefore taken into 348 account. Due to the same reason, it was not possible to examine any potential influence of subtypes of 349 BD on microstate results. 350

Conclusions 351
Our study described altered EEG resting-state microstate temporal parameters in euthymic bipolar 352 patients. Our findings provide an insight into the resting-state global brain network dynamics in BD. 353 Since the increased presence of microstate A is not unique to BD patients, having been reported also 354 in other psychiatric disorders (see Michel and Koenig, 2018), it might be considered only as a non-355 specific electrophysiological marker of BD. Moreover, studies examining possible interactions 356 between microstate dynamics and BD symptoms are needed to better understand the dysfunction of 357 large-scale brain network resting-state dynamics in this affective disorder. 358 6  non-outlier range (whiskers). The x-axis represents the subject group; the y-axis represents the average 402 alpha (8-14 Hz) power across 204 channels. Note significantly decreased alpha power in the BD 403 compared to HC group (p < 0.03, Z value 2.7). 404

9
Conflict of Interest 405 The authors declare that the research was conducted in the absence of any commercial or financial 406 relationships that could be construed as a potential conflict of interest. 407 10 Author Contributions 408 ADdesigned the study, performed the analysis, and wrote the initial draft; JMA, AGD and CPwere 409 responsible for clinical assessment; CMMserved as an advisor; CBcollected the HD-EEG data 410 and was responsible for the overall oversight of the study. All authors revised the manuscript. The funding sources had no role in the design, collection, analysis, or interpretation of the study. 417