Electrical mapping in bipolar disorder patients during the oddballparadigm
Luiza Wanick Di Giorgio Silva h, *, Consuelo Cartier h, Elie Cheniaux l, Fernanda Novis l,Luciana Ang�elica Silveira l, Paola Anaquim Cavaco l, Rafael de Assis da Silva l,Washington Adolfo Batista h, Guaraci Ken Tanaka h, Mariana Gongora a,Juliana Bittencourt e, h, Silmar Teixeira i, Luis Fernando Basile f, g, Henning Budde j, k,Mauricio Cagy d, Pedro Ribeiro a, b, c, Bruna Velasques b, c, h
a Brain Mapping and Sensory Motor Integration of the Federal University of Rio de Janeiro (UFRJ), Brazilb Bioscience Department, School of Physical Education of the Federal University of Rio de Janeiro (EEFD/UFRJ), Rio de Janeiro, Brazilc Institute of Applied Neuroscience (INA), Rio de Janeiro, Brazild Biomedical Engineering Program, COPPE, Federal University of Rio de Janeiro, Rio de Janeiro, Brazile Laboratory of Physical therapy e Veiga de Almeida University of Rio de Janeiro (UVA/RJ), Rio de Janeiro, Brazilf Laboratory of Psychophysiology, Faculdade da Saúde, UMESP, S~ao Paulo, Brazilg Division of Neurosurgery, University of S~ao Paulo Medical School, S~ao Paulo, Brazilh Neurophysiology and Neuropsychology of Attention, Institute of Psychiatry of the Federal University of Rio de Janeiro (IPUB/UFRJ), Rio de Janeiro e RJ,Brazili Brain Mapping and Functionality Laboratory, Federal University of Piauí, Piauí, Brazilj Faculty of Human Sciences, Medical School Hamburg, Hamburg, Germanyk Sport Science, Reykjavik University, Reykjavik, Icelandl Anxiety & Depression Laboratory, Institute of Psychiatry of Federal University of Rio de Janeiro (IPUB/UFRJ), Brazil
a r t i c l e i n f o
Article history:Received 1 June 2015Received in revised form13 October 2015Accepted 15 October 2015
Bipolar disorder (BD) is characterized by an alternated occurrence between acute mania episodes anddepression or remission moments. The objective of this study is to analyze the information processingchanges in BP (Bipolar Patients) (euthymia, depression and mania) during the oddball paradigm, focusingon the P300 component, an electric potential of the cerebral cortex generated in response to externalsensorial stimuli, which involves more complex neurophysiological processes related to stimulusinterpretation. Twenty-eight bipolar disorder patients (BP) (17 women and 11 men with average age of32.5, SD: 9.5) and eleven healthy controls (HC) (7 women and 4 men with average age of 29.78, SD: 6.89)were enrolled in this study. The bipolar patients were divided into 3 major groups (i.e., euthymic,depressive and maniac) according to the score on the Clinical Global Impression e Bipolar Version (CGI-BP). The subjects performed the oddball paradigm simultaneously to the EEG record. EEG data were alsorecorded before and after the execution of the task. A one-way ANOVA was applied to compare the P300component among the groups. After observing P300 and the subcomponents P3a and P3b, a similarity ofamplitude and latency between euthymic and depressive patients was observed, as well as smallamplitude in the pre-frontal cortex and reduced P3a response. This can be evidence of impaired infor-mation processing, cognitive flexibility, working memory, executive functions and ability to shift theattention and processing to the target and away from distracting stimuli in BD. Such neuropsychologicalimpairments are related to different BD symptoms, which should be known and considered, in order todevelop effective clinical treatment strategies.
order patients; EEG, electroencephalogram; ERP, event-related potential; GABA, gamma-aminobutyric acid; HC, healthychizophrenic patients.ry Motor Integration Laboratory, Institute of Psychiatry of the Federal University of Rio de Janeiro (IPUB/UFRJ), Av.60, Brazil.i Giorgio Silva).
L.W. Di Giorgio Silva et al. / Journal of Psychiatric Research 72 (2016) 64e71 65
1. Introduction (2007, 2012), sensory gating deficit has been proposed as anendophenotype for BD and schizophrenia. Early auditory gamma
Bipolar disorder (BD) and its neurological, cognitive andbehavioral bases have beenwidely investigated. BD is considered tobe a relatively frequent and chronic psychiatric condition, causingprofessional and social difficulties or incapacitation (Akiskal et al.,2000; Fleck et al., 2003; Hisatugo et al., 2009; Kaplan et al., 1997).According to €Ozerdem et al. (2008), BD involves various cognitivedysfunctions, even in the euthymic phase of the illness. Emotionalderegulation and cognitive deficits in euthymia are indicators of anenduring pathology in BD. Disruptions of the connections amongthe frontal cortex, amygdala, basal ganglia, thalamus, entorhinalcortex and hippocampus are probable participants in the underly-ing pathology of BD (Atagün et al., 2013; Dupont et al., 1995;Blumberg et al., 2002; Caligiuri et al., 2004; Phillips et al., 2003;Strakowski et al., 2005). These connections are also believed toserve in the modulation of cognition and emotional consonance(Strakowski et al., 2005).
Previous studies have demonstrated cognitive deficits specificfor each mood phase of bipolar disorder. €Ozerdem et al. (2008)identified manic patients to display signs of dysfunction in atten-tional measures, complex processing and memory. Having an acuteepisode of mania or depression is suggested to cause damage to thelearning and memory systems (Bearden et al., 2001). Recently, newevidence corroborates the hypothesis of inflammation and neuro-degeneration in BD and the relation between number of moodepisodes and neurocognitive dysfunction. The understanding of theneurobiology and neuroimaging of BD progression and activitycontributes to the establishment of BD biomarkers, which includeinflammatory cytokines, neurotrophins, mitochondrial dysfunc-tion, oxidative stress, epigenetic effects, and morphometric andneurostructural abnormalities. These parameters appear to besensitive to the illness stage, and they are indeed the firstbiochemical indicators of the staging model in BD (McGorry et al.,2006; Berk et al., 2007; Berk at al., 2011; Roda et al., 2015). Thisway, we suggest P300 components to be investigated as a BDbiomarker, specifically related to information processing andcognitive deficits. According to Purcell et al. (1997), informationprocessing is associated with neurocognitive dysfunctions of BD.Attentional and cognitive alterations are significant in different BDstates and they also persist into euthymic phases (Maekawa et al.,2013). Cognitive deficits in response inhibition, verbal memoryand attention persist across mood phases but are enhanced duringthe manic and depressive states (Robinson et al., 2006). These pa-tients present deficits in a broad range of cognitive functions, suchas verbal memory, sustained attention, executive function aspectsand emotional processing (Andersson et al., 2008; Maekawa et al.,2013). Comparative studies between bipolar disorder patients (BP)and depressive patients showed that, during the manic state, it isharder to maintain attention and inhibit inadequate behaviors,while during depression the problem is shared attention (Murphyet al., 1999).
Electroencephalography (EEG) has been used with BP to verifyrhythmic changes in brain functions in both depression and maniastates (Atagün et al., 2013; Cole et al., 1993; El-Badri et al., 2001).This kind of electrophysiological research is important to identifyresulting changes in cognitive dysfunctions, especially attention.However, most of the studies investigating electrocortical changesin BP used different methodology and data processing. Maekawaet al. (2013) used electrical mapping to show that BP have deficitsin visual information processing, starting from the very early stagesall the way to higher-level cognitive functions. The EEG study ofYeap et al. (2009) also evidenced that visual processing deficit isapparent in schizophrenic patients and BP. According to Hall et al.
band response has been used to assess basic brain functions asso-ciated with auditory perception and showed that BP and schizo-phrenic patients featured reduced early evoked gamma bandresponse (Hall et al., 2011; Roach and Mathalon, 2008). Donchinand Coles (1988) used P2 and P3 ERP (event related potential)components to assess higher-order cognitive processes relatedwith attention, working memory and information processingspeed. Patients with both disorders showed impaired central P3ERPs, but P2 ERP deficit has been documented only in schizo-phrenic patients (O'Donnell et al., 2004). Although Donchin andColes (1988) observed the P3 component, there is a lack ofstudies investigating more deeply the P300 component in the BPstates, and among BP and healthy subjects. The objective of thisstudy is to analyze the alterations of information processing in BP(euthymia, depression and mania phases) through the observationof the P300 component. We hypothesized BP to present a delay ininformation processing, represented by a higher P300 latency,mainly in the depressive group. On the other hand, we expect tofind lower P300 amplitude for BP.
2. Materials and methods
2.1. Subjects
Eleven healthy controls (7 women and 4 men with average ageof 29.78, SD: 6.89) and twenty-eight bipolar patients (17 womenand 11 men with average age of 32.5, SD: 9.5) under treatmentparticipated of this study. Patients with a lifetime history of BD andthose with only current history of BD were included. They werediagnosed according to the DSM-IV (Diagnostic and StatisticalManual of Psychiatric Disorders-fourth edition) (AmericanPsychiatric Association, 1994), and they were asked to suspendmedication one day before the exam. Patients with comorbiditieswere excluded from the study. The subjects were recruited from thePsychiatry Institute of the Federal University of Rio de Janeiro andboth patients and controls were interviewed using the SCID-I(Structured Interview for DSM-IV) (First et al., 1996). All partici-pants had normal or corrected-to normal vision and no sensory,motor, cognitive or attentional deficits. Volunteers who proved tohave no present or past psychiatric condition and to be medicallyhealthy upon physical examinationwere considered for the controlgroup. All patients provided written informed consent beforeentering the study, according to the Declaration of Helsinki. Theexperiment was approved by the Ethics Committee of the Psychi-atric Institute of the Federal University of Rio de Janeiro (IPUB/UFRJ). According to their score on the Clinical Global Impression e
Bipolar Version (CGI-BP) (Spearing et al., 1997) on the day of theexperiment, bipolar patients were divided into 3 major groups:euthymic (n ¼ 10), depressive (n ¼ 8) and manic (n ¼ 10).
2.2. Tasks and procedures
The subjects performed the task in a sound and light-attenuatedroom, in order to minimize sensory interference. The volunteersseated in front of a 1500 monitor. First, EEG data was collected at restfor each subject during three minutes. After this, the subjectsexecuted the Oddball Paradigm (explained below) simultaneouslyto the EEG record, and three more minutes of EEG at rest wererecorded. The Oddball paradigm consists of two stimuli presentedrandomly, with one of them occurring relatively infrequently. Thesubjects need to discriminate target (infrequent) from non-targetor standard stimuli (frequent). In the present experiment, target
L.W. Di Giorgio Silva et al. / Journal of Psychiatric Research 72 (2016) 64e7166
stimuli corresponded to a square and non-target stimuli to a circle.Subjects were instructed to respond as quickly as possible to thetarget stimulus by pressing a button on a joystick (Model QuickShot- Crystal CS4281). Each stimulus lasted 2.5 s, being this thesame interval time between stimuli, with the screen turned off. Thevisual stimulus was presented on the monitor by the event-relatedpotential (ERP) acquisition software, developed in Delphi 5.0(Inprise Co.). The acquisition software recorded event-related po-tentials for the F3, Fz, F4, C3, Cz C4, P3, Pz and P4 electrode sites.P300 is the greatest positive-going peak amplitude of thewaveformwithin a time window of 250e500 ms, in relation to a pre-stimulusbaseline. The baselinewas defined as themean voltage over 120msbefore the onset of the stimulus. Each subject was submitted to sixblocks of 10 trials. In other words, the square was presented 10times in each block.
2.3. EEG data acquisition
The EEG signal acquisition was recorded using the 20-channelBraintech 3000 (EMSA) EEG system, together with the ERP Acqui-sition program already described. This program was employed tofilter the data: Notch (60 Hz), high-pass of 0.3 Hz and low-pass of25 Hz (order 2 Butterworth). Twenty-one electrodes were arrangedon a lycra cap (Eletro Cap Inc., Fairfax, VA) along the scalp on thefrontal, temporal, parietal and occipital areas, according to the 10/20 system protocol, and two more electrodes were positioned onthe earlobes, set as a reference point, yielding 20 mono-pole deri-vations to them (using Fpz as ground electrode). The caps wereindividually adjusted and put on each subject, according to eachindividual's circumference and anatomy proportions. The signalcorrespondent to each EEG derivation resulted from the electricpotential difference between each electrode and the pre-established reference (earlobes).
First, the impedance levels of each electrode were calculated,and they were kept below 10 kU. The ocular electric activity wasestimated by attaching two 9-mm-diameter electrodes in a bipolarmontage. The electrodes were positioned, respectively, above andbelow the right eye orbit, in order to register vertical ocularmovements, and on the external corner of the same eye, in order toregister horizontal ocular movements. Visual artifacts were a prioriinspected through a data visualization program using the Matlab5.3® (The Mathworks, Inc.).
2.4. Data processing and analysis
The electroencephalographic signals collected during theexperiment were processed using methods developed by the BrainMapping and Sensorimotor Integration Laboratory of the Psychia-try Institute of the Federal University of Rio de Janeiro in a Matlab5.3® environment. Visual inspection and Independent ComponentAnalysis (ICA) were applied to quantify reference-free data byremoving possible sources of task-induced artifacts. Data from in-dividual electrodes exhibiting loss of contact with the scalp or highimpedances (>10 kU) were deleted, as were data from single-trialepochs that exhibited excessive movement artifact (±100 mV). ICAwas then applied to identify and remove any artifacts that remainedafter the initial visual inspection. ICA is an information maximiza-tion algorithm that derives spatial filters by blind source separationof the EEG signals into temporally independent and spatially fixedcomponents. Independent components resembling an eye blink ormuscle artifact were removed, and the remaining componentswere then projected back onto the scalp electrodes by multiplyingthe input data by the inverse matrix of the spatial filter coefficientsderived from ICA, using established procedures. The ICA-filtereddata were then re-inspected for residual artifacts, using the
rejection criteria described above. Epochs were selected between0.5 s before and 1.5 s after the stimulus. The total number of epochsused after visual inspection and ICA for each group was as follows:Healthy controls (n ¼ 621); euthymic group (n ¼ 998); depressiongroup (n ¼ 449); and manic group (n ¼ 560).
2.5. Timeefrequency analysis
ERPs transform was computed for the Cz and Pz electrodes,since P300 is more prominent for these electrodes. Time-FrequencyAnalysis was plotted using the EEGLAB toolbox (Delorme andMakeig, 2004) as a qualitative analysis. It was used to visualizeand compare low frequencies, such as delta (0.3e4 Hz) and theta(4e8 Hz), related to time arising from the 4 experimental groups(i.e., control, euthymic, depressive, manic).
2.6. Statistical analysis
A one-way ANOVA (SPSS version 18) was applied, in order toinvestigate the factor group (i.e., HC, euthymic, depressive andmaniac) for P300 and reaction time, separately. Significant differ-ence was set at p < 0.05.
3. Results
3.1. reaction time
The one-way ANOVA demonstrated a main effect for group(p < 0.05). The post-hoc analysis showed no difference betweeneuthymic (average: 451.57 ms, SD: 77.47 ms) and depression(average: 455.94 ms, SD: 81.82 ms) groups. We also observed thatthe control (average: 414.3109 ms, SD: 79.33134) and manic(average: 433.0641 ms, SD: 81.86123) groups differed from theother groups (Fig. 1).
3.2. Event-related potentials
P300 amplitude and latency were observed, as well as thesubcomponents P3a and P3b. Results will be explained and pre-sented according to specific regions: frontal (F3, F4 and Fz), central(C3, C4 and Cz) and parietal (P3, P4 and PZ).
3.2.1. Frontal areaThe one-way ANOVA demonstrated a main effect for group for
the Fz, F4 and F3 electrodes for P300 amplitude and latency(p < 0.05). For the F4 (Fig. 3b) and Fz (Fig. 2a) electrodes, differencewas found between the control [(F4 e amp: 0.5733 mV; lat: 335 ms)(Fz - amp: 0.6139 mV; lat: 330 ms)] and manic [(F4 e amp:0.1443 mV; lat: 350) (Fz e amp: 0.2824 mV; lat: 345)] groups. Nodifference was found between euthymic [(F4 - amp: 0.4792 mV; lat:370 ms) (Fz e amp: 0.5396 mV; lat: 360 ms)] and depressive [(F4 -amp: 0.487 mV; lat: 375 ms) (Fz e amp: 0.5419 mV; lat: 355 ms)]groups. When analyzing the F3 electrode, difference was observedamong the four groups, i.e., control (amp: 0.31 mV; lat: 330 ms),euthymic (amp: 0.3908 mV; lat: 365 ms), depressive (amp: 0.3393mV/lat: 365 ms) and manic (amp: 0.08819 mV; lat: 350 ms) (Fig. 3a).
3.2.2. Central areaThe one-way ANOVA demonstrated a main effect for group for
the Cz and C4 electrodes for P300 amplitude and latency (p < 0.05).P3a and P3b components were observed for the Cz and C4 elec-trodes in HC. For the Cz electrode (Fig. 2b), no difference was foundin the latency among euthymic (amp: 0.4737 mV; lat: 355 ms),depression (amp: 0.4884 mV; lat: 355 ms) and manic groups (amp:0.4411 mV; lat: 355 ms); however amplitude was higher for the
Fig. 1. Comparison among control, euthymic, depression and manic groups for reac-tion time. Data represent mean and SD for Reaction Time. A one-way ANOVA showedsignificant difference in the marked bars (* represents significant p values).
L.W. Di Giorgio Silva et al. / Journal of Psychiatric Research 72 (2016) 64e71 67
manic group. The control group (amp: 0.6119 mV; lat: 335 ms)differed from the other groups in amplitude and latency. For the C4electrode, all groups were different among them: control (amp:0.5148 mV; lat: 350 ms), euthymic (amp: 0.581 mV; lat: 365 ms),depressive (amp: 0.5766 mV; lat: 375 ms) and manic (amp:0.3941 mV; lat: 375 ms) (Fig. 3c). No difference was found amongthe groups for the C3 electrode (p > 0.05).
3.2.3. Parietal areaThe one-way ANOVA demonstrated a main effect for group for
the Pz and P4 electrodes for P300 amplitude and latency (p < 0.05).P3a and P3b components were observed for the Pz electrode in theeuthymic and depressive groups. Significant difference was foundamong all the groups for the Pz electrode: control (amp: 1.107 mV;lat: 480 ms); euthymic (amp: 0.3998 mV; lat: 395 ms); depressive(amp: 0.3674 mV; lat: 410 ms) and manic (amp: 0.6308 mV; lat:395 ms) (Fig. 2c). For the P4 electrode, difference was found amongall groups for the P300 latency; for P300 amplitude, the controlgroup (amp: 0.6326 mV; lat: 470 ms) was different from the others,and no difference was observed among the mood phases of bipolardisorder [euthymic (amp: 0.5234 mV; lat: 390 ms), depressive(amp: 0.4905 mV; lat: 400 ms) and manic (amp: 0.4526 mV; lat:385 ms)] (Fig. 3d). No significant difference was found among thegroups for the P3 electrode (p > 0.05).
3.2.4. Timeefrequency analysisFig. 4 shows Time-Frequency Analysis plots for all the groups
investigated in the medial central area (i.e., Cz electrode) (Fig. 4a)and medial parietal area (i.e., Pz electrode) (Fig. 4b). The qualitativedata of the Time-Frequency Analysis show changes in the spectral
Fig. 2. Comparison among control, euthymic, depression and manic groups for P300 compoand manic groups for P300 latency and amplitude for the Fz electrode. (b) A one-way ANOVlatency and amplitude for the Cz electrode. (c) A one-way ANOVA showed significant diffe
power for all groups, with a more explicit difference betweenhealthy controls and bipolar patients.
4. Discussion
The purpose of this study was to investigate neurophysiologicaldifferences and similarities between BP and HC, through the P300analysis. Our main findings were higher P300 amplitude in HC, lowP300 amplitude in manic patients, similar amplitude and latency ineuthymic and depressive patients and slower reactivity of BP. P300amplitude and latency in BP during both depressive and euthymicperiods did not differ in many areas.
P300 latency indicates stimulus processing time, largely inde-pendent of behavioral response selection and execution (Duncanand Donchin, 1982). We observed that BP demonstrated a delayin the information processing, represented by prolonged P300 la-tency. Before the different phases of information processing(acquisition, primary and secondary analysis, decision and execu-tion), BP have been observed to feature a delay on decision makingand execution moments. This result is in agreement with theresearch by Schulze et al. (2008) that found BP to have P300 latencydelays compared to controls at all recording sites, with the mostpronounced differences at parietal and central sites. These authorsalso indicated delayed P300 latency at midline sites to be associatedwith familial risk for psychotic BD.
Cognitive demands during task processing have an influence onP300 (Muir et al., 1991; Salisbury et al., 1999; Souza et al., 1995;Schulze et al., 2008; O'Donnell, 2004; Pierson et al., 2000). HCpresented higher amplitude than BP for the Fz, Cz and Pz elec-trodes, which are typically responsible electrodes for P300 poten-tial. Lower P300 amplitude for the BD group, when compared to HC,can be associated with neurocognitive dysfunction. We proposelower P300 amplitude to be considered a biomarker for BD thatunderlies the neurocognitive deficits. Specifically, dysfunctions inattention, memory and executive function are well documented inBD (Azorin et al., 1995; Bearden et al., 2001; Clark et al., 2002,2005). However, most of these studies did not observe the eventrelated potentials (ERPs). Previous studies observing P300 ampli-tude in healthy subjects related lower P300 amplitude withreduced attention allocation during a task. More symptomaticsamples may be more likely to demonstrate amplitude deficits(Schulze et al., 2008). P300 amplitude reduction in BP is in agree-ment with previous paper results (Salisbury et al., 1999; O'Donnellet al., 2004). The manic group showed small P300 amplitude,mainly for the frontal electrodes (F3, Fz and F4), which reflect theactivity of Brodmann cortical area 8. This region is located justbefore the pre-motor cortex and it participates in executive controland behavior, inductive reasoning, planning, memory processesand working memory (Trans Cranial Technologies, 2012).
According to Gordeev (2008), P300 amplitude is significantly
nent (p < 0.05). (a) A one-way ANOVA showed significant difference between controlA showed significant difference between control group and the other groups for P300rence among all groups for P300 latency and amplitude for the Pz electrode.
Fig. 3. Comparison among control, euthymic, depression and manic groups for P300 component (p < 0.05). (a) A one-way ANOVA showed significant difference among all groupsfor P300 latency and amplitude for the F3 electrode. (b) A one-way ANOVA showed significant difference between control and manic groups for P300 latency and amplitude for theF4 electrode. (c) A one-way ANOVA showed significant difference among all groups for P300 latency and amplitude for the C4 electrode. (d) A one-way ANOVA showed significantdifference among all groups for P300 latency and difference between control group and bipolar patient group for P300 amplitude for the P4 electrode.
L.W. Di Giorgio Silva et al. / Journal of Psychiatric Research 72 (2016) 64e7168
influenced by the complexity of a stimulus, while P300 latency isdirectly related to the speed at which the task is executed. Severalresearchers correlate changes in P300 amplitude with changes inthe level of attention, pointing it as being directly proportional tothe level of attention in the execution of a task (Gordeev, 2008). Thepre-frontal cortex is related to executive functions, cognitive flex-ibility, working memory (De Carvalho et al., 2010), behavioralplanning and complex thoughts, such as decisionmaking, attentioncontrol (Bechara et al., 1997; Damasio, 1994), behavior modulation(Windmann et al., 2002; Waltz et al., 1999), and emotional regu-lation (Windmann et al., 2002; Lobo et al., 2011).
This small amplitude can be related to an impairment of thesecognitive functions in BP during the manic state. This fact is inagreement with the mania symptoms, since they include superfi-cial attention, disorganized thought, quantitative alterations ofperception, distractibility, impulsive behavior, among others (Clarket al., 2002; Thompson et al., 2005; Goldberg, 2010). €Ozerdem et al.(2008) pointed out that manic patients also display signs ofdysfunction in attentional measures, complex processing andmemory. Clark et al. (2005) concluded that BP feature a deficit insustained attention during the acute manic crisis. The frontal re-ductions in BP may reflect abnormalities in a hypothetical frontalgenerator, consonant with reports about altered frontal lobe func-tioning in mania (Salisbury et al., 1999).
El-Badri et al. (2001) showed young euthymic patients withbipolar affective disorder to feature significant EEG abnormalitiesand cognitive impairments, as well as disturbed EEG activity at rest,when compared with control subjects of similar age. This reductionin P300 amplitude can also be seen in schizophrenic patients (SP).According to Bestelmeyer (2012), BP and SP could not be differen-tiated based on their ERPs. If the P300 amplitude reflects
attentional resource allocation, SP allocate more resources to thedistracting task-irrelevant stimuli than to the task-relevant stimuli(Grillon et al., 1990). Previous research has shown that BP showsome attentional deficits, which are, however, not as severe as theones found in schizophrenia (Bozikas et al., 2005).
Results also identified euthymic and depressive patients topresent marked P3a and P3b components in the parietal area. In thecentral area, these elements (i.e., P3a and P3b) were seen only inthe HC group, whose amplitudewas higher than for BP. Grillon et al.(1990) demonstrated SP to show smaller P3a and P3b amplitudescompared to HC. These authors suggested HC and patients withschizophrenia to process target and distracting stimuli differently.The reduced P3a response in the patients with bipolar disordersuggests an impaired covert orienting response or an inability toshift the attention to meaningful auditory stimuli (Friedman et al.,2001). Bestelmeyer (2012) compared the P300 of BP, SP and HC andfound P3a to be slightly greater than P3b in all groups. According tothe research conducted by Jahshan (2012), BP exhibited large P3areductions for Fz, compared to the HC group, and medium re-ductions compared to the schizophrenia group. This finding sug-gests that both groups of patients may have problems detectingchanges in their auditory environment.
The behavioral analysis of reaction time confirmed the electro-physiological findings already described, showing that BP haveimpaired information processing, which is slower, when comparedto HC. Euthymic and depressive patients presented slower reac-tivity to the stimuli, as was also shown by other studies (Kertzmanet al., 2010; Lampe et al., 2004; Pier et al., 2004), which highlight aslower response of BP during reaction time tasks that can be relatedto a dysfunction of information processing during the depressivestate (Azorin et al., 1995; Rose and Ebmeier, 2006).
Fig. 4. TimeeFrequency Analysis plots for all the groups investigated in the medial central area (i.e., Cz electrode) (Fig. 4a) and medial parietal area (i.e., Pz electrode) (Fig. 4b). Thequalitative data of the TimeeFrequency Analysis show changes in the spectral power for all groups, with a more explicit difference between healthy controls and bipolar patients.
L.W. Di Giorgio Silva et al. / Journal of Psychiatric Research 72 (2016) 64e71 69
5. Conclusion
The main findings of this study were: higher P300 amplitude inHC, low P300 amplitude in manic patients and similar amplitudeand latency in euthymic and depressive patients. The manic groupshowed the smallest P300 amplitude, mainly for the frontal elec-trodes, which can be related to manic state symptoms. This studyprovided some evidence of cognitive deficits in BP, like a delay ininformation processing and reduced attention allocation, which isin agreement with previous studies about BD cognitive aspects.Such finding is clinically relevant, since these neuropsychologicalimpairments are related to different BD symptoms, which shouldbe known and considered, in order to improve treatment strategies.The novelty of this research is the electrophysiological analysis ofP300, in order to differentiate ERP aspects at different moments ofBD. The small amplitude for the pre-frontal area electrodes can berelated to an impairment of cognitive flexibility, executive func-tions, working memory and other cognitive functions. The reducedP3a response suggests an impaired ability to shift the attention andprocessing to the target and away from the distracting stimuli in BP.As future directions, we suggest the execution of more studies us-ing the same methodology of P300 components. This methodo-logical pattern could contribute to strengthen our results andreinforce the importance of P300 as a biomarker for BD. We alsopropose that new studies comparing the EEG parameters acrossgroups depending on the number of experienced mood episodes
should be implemented, in order to relate more closely the elec-trocortical changes in BD using the new inflammation and neuro-progression theories in BD.
Conflicts of interest
The authors and the represented institutions confirm that thecontent of the present article has no conflict of interest.
Funding Source
There were no funding source. The financial investment wasdone by the authors.
Acknowledgments
The authorswish to acknowledge grant support from the Instituteof Psychiatry of the Federal University of Rio de Janeiro (Rio deJaneiro, Brazil), from theLaboratoryof Panic andRespiration and fromthe Institute of Psychiatry and the Brain Mapping and SensoryMotorIntegration Laboratory of the Federal University of Rio de Janeiro.
References
Akiskal, H.S., Bourgeois, M.I., Angst, J., M€oller, H.J., Hirschifild, R., 2000. Re-evalu-ating the prevalence of and diagnostic comparison within the broad clinical
logical and electropy- siological indices of neurocognitive dysfunction in bi-polar ii disorder. Bipolar Disord. 10, 888e899.
Atagün, M.I., Güntekin, B., €Ozerdem, A., Tülay, E., Basar, E., 2013. Decrease of thetaresponse in euthymic bipolar patients during an oddball paradigm. Cogn.Neurodyn. 7, 213e223.
Azorin, J.M., Benhaim, P., Hasbroucq, T., Possamai, C.A., 1995. Stimulus preprocess-ing and response selection in depression: a reaction time study. Acta Psychol.89, 95e100.
Bearden, C.E., Hoffman, K.M., Cannon, T.D., 2001. The neuropsychology andneuroanatomy of bipolar affective disorder: a critical review. Bipolar Disord. 3(3), 106e150 discussion 151e153.
Bechara, A., Damasio, H., Tranel, D., Damasio, A.R., 1997. Deciding advantageouslybefore knowing the advantageous strategy. Science 275, 1293e1295.
Berk, M., Copolov, D.L., Dean, O., Lu, K., Jeavons, S., Shapkaitz, I., et al., 2007. Anti-oxidant treatment of the glutathione deficiency in bipolar disorder with N-acetylcysteine: a double-blind randomised placebo controlled trial. BipolarDisord. 9, 8e9.
Berk, M., Kapczinski, F., Andreazza, A.C., Dean, O.M., Giorlando, F., Maes, M., et al.,2011. Pathways underlying neuroprogression in bipolar disorder: focus oninflammation, oxidative stress and neurotrophic factors. Neurosci. BiobehavRev. 35 (3), 804e817.
Bestelmeyer, P.E.G., 2012. The visual P3a in schizophrenia and bipolar disorder:effects of target and distractor stimuli on the P300. Psychiatry Res. 197,140e144.
Bozikas, V.P., Andreou, C., Giannakou, M., Tonia, T., Anezoulaki, D., Karavatos, A.,et al., 2005. Deficits in sustained attention in schizophrenia but not in bipolardisorder. Schizophr. Res. 78, 225e233.
Caligiuri, M.P., Brown, G.G., Meloy, M.J., Eyler, L.T., Kindermann, S.S., Eberson, S.,et al., 2004. A functional magnetic resonance imaging study of cortical asym-metry in bipolar disorder. Bipolar Disord. 6, 183e196.
Clark, L., Kempton, M.J., Scarn�a, A., Grasby, M., Goodwin, G.M., 2005. Sustainedattention-deficit confirmed in euthymic bipolar disorder but not in first-degreerelatives of bipolar patients or euthymic unipolar depression. Biol. Psychiatry 57(2), 183e187.
Damasio, A., 1994. Descartes' Error. Emotion, Reasoning and the Human Brain. AvonBooks, New York.
De Carvalho, M.R., Dias, G.P., Cosci, F., de-Melo-Neto, V.L., Bevilaqua, M.C.,Gardino, P.F., et al., 2010. Current findings of fMRI in panic disorder: contribu-tions for the fear neurocircuitry and CBT effects. Expert Rev. Neurother. 10,291e303.
Delorme, A., Makeig, S., 2004. EEGLAB: an open source toolbox for analysis ofsingle-trial EEG dynamics including independent component analysis. J. Neu-rosci. Methods 134, 9e21.
Donchin, E., Coles, M.G.H., 1988. Is the P300 component a manifestation of contextupdating. Behav. Brain Sci. 11, 357e374.
Duncan-Johnson, C.C., Donchin, E., 1982. The P300 component of the event-related brain potential as an index of information processing. Biol. Psychol.14, 1e52.
Dupont, R.M., Jernigan, T.L., Heindel, W., Butters, N., Shafer, K., Wilson, T.,Hesselink, J., et al., 1995. Magnetic resonance imaging and mood disorders.Localization of white matter and other subcortical abnormalities. Arch. Gen.Psychiatry 52, 747e755.
El-Badri, S.M., Ashton, C.H., Moore, P.B., Marsh, V.R., Ferrier, I.N., 2001. Electro-physiological and cognitive function in young euthymic patients with bipolaraffective disorder. Bipolar Disord. 3 (2), 79e87.
First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1996. Structured Clinical Inter-view for DSM-IV Axis I Disorders, Clinician Version. American Psychiatric Press,Inc, Washington, DC.
Fleck, M.P.A., Lafer, B., Sougey, E.B., Del Porto, J.A., Brasil, M.A., Juruena, M.F., 2003.Diretrizes da Associaç~ao M�edica Brasileira para o tratamento da depress~ao. Rev.Bras. Psiquiatr. 25 (2), 114e122.
Friedman, D., Cycowicz, Y.M., Gaeta, H., 2001. The novelty P3: an event-related brainpotential (ERP) sign of the brains evaluation of novelty. Neurosci. Biobehav Rev.25, 355e373.
Goldberg, J.F., 2010. Ruling in and ruling out “caseness” in the bipolar spectrum:implications for initial screening. Rev. Bipolar Disord. 12 (5), 539e540.
Gordeev, S.A., 2008. Cognitive functions and the state of nonspecific brain systemsin panic disorders. Neurosci. Behav. Physiol. 38 (7), 707e714. Sep.
Grillon, C., Courchesne, E., Ameli, R., Geyer, M.A., Braff, D.L., 1990. Increaseddistractibility in schizophrenic patients. Electrophysiologic and behavioral ev-idence. Arch. Gen. Psychiatry 47, 171e179.
Hall, M.H., Rijsdijk, F.V., Picchioni, M., Schulze, K., Ettinger, U., Toulopoulou, T., et al.,2007. Substantial shared genetic influences on schizophrenia and event- relatedpotentials. Am. J. Psychiatry 164, 804e812.
Hall, M.H., Spencer, K.M., Schulze, K., McDonald, C., Kalidindi, S., Kravariti, E., et al.,
2011. The genetic and environmental influences of event-related gamma os-cillations on bipolar disorder. Bipolar Disord. 13, 260e271.
Hall, M.-H., Smoller, J.W., Cook, N.R., Schulze, K., Lee, P.H., Taylor, G., et al., 2012December 30. Patterns of deficits in brain function in bipolar disorderand schizophrenia: a cluster analytic study. Psychiatry Res. 200 (2e3),272e280.
Hisatugo, C.L.C., Yazigi, L., Del Porto, J.A., 2009. Cogniç~ao, Afeto e RelacionamentoInterpessoal de Mulheres com Transtorno Afetivo Bipolar. Psicol. Reflex. Crit. 22(3), 377e385.
Jahshan, C., Wynn, J.K., Mathis, K.I., Altshuler, L.L., Glahn, D.C., Green, M.F., 2012.Cross-diagnostic comparison of duration mismatch negativity and P3a in bi-polar disorder and schizophrenia. Bipolar Disord. 14, 239e248.
Kaplan, I.H., Sadock, J.B., Grebb, A.J., 1997. Compendio de psiquiatria: Ciencias docomportamento e psiquiatria clínica. Artes M�edicas, Porto Alegre, RS.
Kertzman, S., Reznik, I., Hornik-Lurie, T., Weizman, A., Kotler, M., Amital, D., et al.,2010. Stroop performance in major depression: selective attention impairmentor psychomotor slowness? J. Affect. Disord. 122, 167e173.
Lampe, I.K., Sitskoorn, M.M., Heeren, T.J., 2004. Effects of recurrent major depressivedisorder on behavior and cognitive function in female depressed patients.Psychiatry Res. 125, 73e79.
Lobo, I., Oliveira, L., David, I.A., Pereira, M.G., Volchan, E., Rocha-Rego, V., et al., 2011.The neurobiology of posttraumatic stress disorder: dysfunction in theprefrontal-amygdala circuit? Psychol. Neurosci. 4 (2), 191e203.
Maekawa, T., Katsuki, S., Kishimoto, J., Onitsuka, T., Ogata, K., Yamasaki, T., et al.,2013. Altered visual information processing systems in bipolar disorder: evi-dence from visual MMN and P3. Front. Hum. Neurosci. 7, 403.
McGorry, P.D., Hickie, I.B., Yung, A.R., Pantelis, C., Jackson, H.J., 2006. Clinicalstaging of psychiatric disorders: a heuristic framework for choosing earlier,safer and more effective interventions. Austral. N. Z. J. Psychiatry 40,616e622.
Muir, W.J., St. Clair, D.M., Blackwood, D.H.R., 1991. Long-latency auditory event-related potentials in schizophrenia and in bipolar and unipolar affective dis-order. Psychol. Med. 21, 867e879.
Murphy, F.C., Sahakian, B.J., Rubinsztein, J.S., Michael, A., Rogers, R.D., Robbins, T.W.,et al., 1999. Emotional bias and inhibitory control processes in mania anddepression. Psychol. Med. 29, 1307e1321.
�ODonnell, B.F., Vohs, J.L., Hetrick, W.P., Carroll, C.A., Shekhar, A., 2004. Auditoryevent-related potential abnormalities in bipolar disorder and schizophrenia.Int. J. Psychophysiol. 53, 45e55.
€Ozerdem, A., Güntekin, B., Tunca, Z., Basar, E., 2008. Brain oscillatory responses inpatients with bipolar disorder manic episode before and after valproate treat-ment. Brain Res. 1235, 98e108.
Phillips, M.L., Drevets, W.C., Rauch, S.L., Lane, R.D., 2003. The neurobiology ofemotion perception I: towards an understanding of the neural basis of normalemotion perception. Biol. Psychiatry 54, 504e514.
Pier, M.P., Hulstijn, W., Sabbe, B.G., 2004. Differential patterns of psychomotorfunctioning in unmedicated melancholic and nonmelancholic depressed pa-tients. J. Psychiatr. Res. 38, 425e435.
Pierson, A., Jouvent, R., Quintin, P., Perez-Diaz, F., Leboyer, M., 2000. Informationprocessing deficits in relatives of manic depressive patients. Psychol. Med. 30,545e555.
Purcell, R., Maruff, P., Kyrios, M., Pantelis, C., 1997. Neuropsychological function inyoung patients with unipolar major depression. Psychol. Med. 27 (6),1277e1285.
Roach, B.J., Mathalon, D.H., 2008. Event-related EEG time-frequency analysis: anoverview of measures and an analysis of early gamma band phase locking inschizophrenia. Schizophr. Bull. 34, 907e926.
Robinson, L.J., Thompson, J.M., Gallagher, P., Goswami, U., Young, A.H., Ferrier, I.N.,et al., 2006. A meta-analysis of cognitive deficits in euthymic patients withbipolar disorder. J. Affect. Disord. 93, 105e115.
Roda, A., Chendo, I., Kunz, M., 2015 JaneMar. Biomarkers and staging of bipolardisorder: a systematic review. Trends Psychiatry Psychother. 37 (1), 3e11.
Rose, E.J., Ebmeier, K.P., 2006. Pattern of impaired working memory during majordepression. J. Affect. Disord. 90, 149e161.
Schulze, K.K., Hall, M.-H., McDonald, C., Marshall, N., Walshe, M., Murray, R.M., et al.,2008. Auditory P300 in patients with bipolar disorder and their unaffectedrelatives. Bipolar Disord. 10, 377e386.
Souza, V.B., Muir, W.J., Walker, M.T., Glabus, M.F., Roxborough, H.M., Sharp, C.W.,et al., 1995. Auditory P300 event-related potentials and neuropsychologicalperformance in schizophrenia and bipolar affective disorder. Biol. Psychiatry 37,300e310.
Spearing, M.K., Post, R.M., Leverich, G.S., Brandt, D., Nolen, W., 1997. Modification ofthe clinical global impressions (CGI) scale for use in bipolar illness (BP): theCGI-BP. Psychiatry Res. 73 (3), 159e171.
Strakowski, S.M., DelBello, M.P., Adler, C.M., 2005. The functional neuroanatomy ofbipolar disorder: a review of neuroimaging findings. Mol. Psychiatry 10,105e116.
Thompson, J.M., Gallagher, P., Hughes, J.H., Watson, S., Gray, J.M., Ferrier, I.N., et al.,2005. Neurocognitive impairment in euthymic patients with bipolar affectivedisorder. Br. J. Psychiatry 186, 32e40.
Trans Cranial Technologies. Cortical Functions Reference. Wanchai, Hong Kong:Trans cranial technologies ldt 20pp. 20-21.
L.W. Di Giorgio Silva et al. / Journal of Psychiatric Research 72 (2016) 64e71 71
Waltz, J.A., Knowlton, B.J., Holyoak, K.L., Boone, K.B., Mishkin, F.S., de MenezesSantos, M., et al., 1999. A system for relational reasoning in human prefrontalcortex. Psychol. Sci. 10, 119e125.
Windmann, S., Sakhavat, Z., Kutas, M., 2002. Electrophysiological evidence revealsaffective evaluation deficits early in stimulus processing in patients with panic
