International Review of Psychiatry, October 2013; 25(5): 604–618

EEG biomarkers in major depressive disorder: Discriminative

power and prediction of treatment response

SEBASTIAN OLBRICH1 & MARTIJN ARNS2,3

1Clinic for Psychiatry and Psychotherapy, University Hospital Leipzig, Germany, 2Department of Experimental Psychology,

Utrecht University, Utrecht, Netherlands, and 3Research Institute Brainclinics, Nijmegen, Netherlands
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

Abstract
Major depressive disorder (MDD) has high population prevalence and is associated with substantial impact on quality of
life, not least due to an unsatisfactory time span of sometimes several weeks from initiation of treatment to clinical response.
Therefore extensive research focused on the identification of cost-effective and widely available electroencephalogram
(EEG)-based biomarkers that not only allow distinguishing between patients and healthy controls but also have predictive
value for treatment response for a variety of treatments. In this comprehensive overview on EEG research on MDD,
biomarkers that are either assessed at baseline or during the early course of treatment and are helpful in discriminating
patients from healthy controls and assist in predicting treatment outcome are reviewed, covering recent decades up to now.
Reviewed markers include quantitative EEG (QEEG) measures, connectivity measures, EEG vigilance-based measures,
sleep–EEG-related measures and event-related potentials (ERPs). Further, the value and limitations of these different
markers are discussed. Finally, the need for integrated models of brain function and the necessity for standardized proce-
dures in EEG biomarker research are highlighted to enhance future research in this field.
For personal use only.

Introduction their way into routine prognostic use, in part due to

Major depressive disorder (MDD) is associated with the unavailability of the measures in clinical everyday
high lifetime prevalence, estimated between 13.2– work, low specificity and also due to the heterogene-
16.5% (Volkert et al., 2013). Its social and economic ity of findings. The electroencephalogram (EEG),
burden and, more importantly, its impact on quality of which is already used in many clinics for routine
life and individual suffering of patients has triggered diagnostic purposes, does meet several of the require-
efforts for identification of biomarkers that might help ments for biomarker research, for example wide
to better predict treatment response in MDD to vari- availability and cost-effectiveness. Since the very
ous treatments. According to the definition, a biomarker beginning of EEG research after the first descriptions
should be assessable objectively and provide informa- of human scalp recordings of neuronal activity
tion about physiological or pathological processes or (Berger, 1933), the possible usage in neuropsychiat-
responses to treatment interventions (Atkinson et al., ric disorders has been emphasized, and is under-
2001). They can be measured in order to help diagnose pinned by recent findings of the EEG as one of the
and stage a disorder, to give a prognostic outline or to most heritable biomarkers (De Gennaro et al., 2008),
predict treatment outcome (prognosis). particularly alpha peak frequency (APF), alpha activ-
In recent years, several anatomical, metabolic ity and the event-related potential (ERP) P300 (Van
and physiological aberrations in MDD have been Beijsterveldt & van Baal, 2002). Further, the EEG
reported; for example, decreased cortical volumes in provides a temporal resolution in a time scale of mil-
prefrontal brain areas such as the dorsolateral pre- liseconds, which is the time-frame at which neuronal
frontal cortex or the subgenual gyrus (e.g. Chang, C.C. activity, and especially cognition, takes place. The
et al., 2011), altered connectivity and activity in fron- EEG does not assess a surrogate marker of neuronal
tal and anterior cingulate cortex (ACC) networks activity, for example glucose utilization or blood oxy-
(Fox et al., 2012; Pizzagalli, 2011) or altered inflam- genation, but directly captures on-going electric
matory cytokine and growth factor levels (Schmidt activity from the brain that makes this technique a
et al., 2011). Still, none of these findings have made valuable complementary brain imaging method.

Correspondence: Sebastian Olbrich, Semmelweißstraße 10, 04103 Leipzig, Germany. Tel: 0049-341-9724364. Fax: 0049-341-9724509. Email: Sebastian.
[email protected]

(Received 27 April 2013; accepted 13 June 2013)

ISSN 0954–0261 print/ISSN 1369–1627 online © 2013 Institute of Psychiatry
DOI: 10.3109/09540261.2013.816269
 EEG biomarkers in major depressive disorder 605
Therefore the EEG provides an excellent basis for EEG-based biomarkers for differentiation between
the development of biomarkers, although heteroge- MDD and healthy subjects
neous findings of EEG parameters in MDD, not
Quantitative EEG. EEG data comprises information
least due to different methodologies and a lack of
about neural activity from different brain sites at dif-
standardization, accompanied sometimes by over-
ferent frequencies. Time series analysis methods,
interpretation of findings, have blunted the value of
especially, such as fast Fourier transformation, for
EEG-based measures. Further, the assumption of a
example, allow quantification of cortical activity to
homogeneous biological pathway in MDD, and thus
be assessed via the EEG and to reliably separate dif-
one characteristic neurophysiological signature, is
ferent components such as the power of the main
misleading. Instead, the clinical diagnosis of MDD
frequency bands, delta (1–3 Hz), theta (4–7 Hz),
reflects a cluster of observable behavioural and
alpha (8–12 Hz), and beta (13–25 Hz). Elevated
reported affective alterations with many possible
EEG alpha activity during rest in depressed
underlying biological pathways. To improve and
patients,especially, has been one of the main and
guide future research, the aim of this work is to give
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

most consistent findings from studies of recent

a descriptive and comprehensive overview of EEG-
decades, in line with the first observation from
based research in MDD.
Lemere (1936): ‘The ability to produce “good” alpha
waves seems to be a neurophysiological characteristic
which is related in some way to the affective capacity
Methods
of the individual.’ While Begić et al. (2011), Jaworska
A literature search was performed in the PubMed et al. (2012a), Roemer et al. (1992) and von Knor-
database using the following Keywords: ‘depression’ ring et al. (1983) reported elevated absolute alpha
or ‘major depressive disorder’ and ‘EEG’, ‘QEEG’, power, others described increases in relative power
‘electroencephalogram’. From the obtained results (John et al., 1988; Prichep & John, 1992). Increases
the title and abstract were screened and it was decided were mainly located to parietal and frontal (e.g.
For personal use only.

whether they contributed to the field of EEG bio- Grin-Yatsenko et al., 2009; Jaworska et al., 2012a)
marker research in MDD or not. Due to the immense or occipital sites (Bruder et al., 2008). Recently,
body of literature in this field, this review is not Grin-Yatsenko (2010) replicated these findings in a
intended as a systematic review; rather, it provides a large sample with increased alpha (and theta power)
comprehensive overview of EEG-based findings on in patients in early stages of depression. Other stud-
the road towards more powerful electrophysiological ies failed to find alpha power differences between
biomarkers in MDD. patients and healthy controls (Flor-Henry et al.,
1979; Knott & Lapierre, 1987) or reported decreased
(relative) alpha activity in patients with MDD in
Results
comparison to other patient groups (Pozzi et al.,
A first description relating EEG patterns to affective 1995) or in treatment-resistant patients with MDD
capacity has already been given by Lemere (1936), (Price et al., 2008). As will be reported later, excess
only 5 years after Berger’s first description of the alpha is also associated with a favourable response to
human EEG (Berger, 1933). Since then, different antidepressive treatments (Ulrich et al., 1984) and
measures have emerged and have been quantified to antidepressants decrease alpha power (Itil, 1983),
distinguish between patients with MDD and healthy further supporting a functional role of this rhythm in
subjects. The first part of this review is dedicated to at least a subgroup of MDD patients.
these EEG biomarkers that discriminate between It has been proposed that the effects of the brain-
MDD and healthy subjects. derived neurotrophic factor (BDNF) Val66Met
Because of the known delayed onset of clinical polymorphism on trait depression were mediated by
effect of most antidepressants and their limited effi- EEG alpha power in a study of 305 healthy controls
cacy, it is also desirable to find parameters that help (HCs) (Gatt et al., 2008), which was recently repli-
to separate responders from non-responders before cated in MDD patients (Zoon et al., 2013), where
or soon after treatment initiation. Hence, the second the BDNF MetMet polymorphism in MDD seems
part of this review covers biomarkers that are assessed to be associated with low-voltage alpha EEG.
either at baseline and/or at the early stages of treat- Besides alpha power measures, increased slow
ment (referred to as ‘treatment emergent’ biomark- wave activity in MDD has also been reported (Kwon
ers) and provide a predictive value for clinical et al., 1996; Lieber & Prichep, 1988; Nyström et al.,
outcome of treatment. Table 1 gives a brief overview 1986; Roemer et al., 1992). Interestingly, several
of the most important measures for discrimination prognostic studies have reported the excess theta
between patients with MDD and healthy subjects group in MDD to be related to non-response to anti-
and for treatment prediction in MDD. depressant treatments (Iosifescu et al., 2009; Knott
 Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14
For personal use only.

606

Table 1. Listing of EEG-based methods for discrimination between patients with major depressive disorder (MDD) and healthy controls, and for prediction of treatment outcome.

Method Measures Usage Main findings in MDD

qEEG E.g. EEG power via FFT Quantification of EEG power at different Increased EEG alpha activity at especially occipital sites in
EEG frequency bands MDD; frontal widespread increases of slow activity are
associated with non-response
EEG source estimates, e.g. Current source FFT density Estimation of the intracortical generators of Local increases of theta activity within the anterior
LORETA scalp EEG activity cingulate cortex are associated with good treatment
S. Olbrich & M. Arns

response
Alpha asymmetry Left/right hemispheric alpha power Computation of EEG power differences Decreased right prefrontal alpha activity and an increased
between, e.g. hemispheres left prefrontal alpha activity in MDD have been
discussed controversially
EEG connectivity, (non-linear Quantified coupling of two EEG time Assessment of (phase and/or amplitude) Altered connectivity patterns have been reported in MDD;
and linear) coherence series synchronization of neural activity between further characterization is needed
brain areas
EEG vigilance Different vigilance stages from Assessment of declines of vigilance, i.e. tonic A hyper-stable regulation with fewer declines to lower
wakefulness to sleep onset brain arousal during eyes-closed resting EEG vigilance stages is found in MDD
condition
EEG cordance A complex measure that comprises Assessment of EEG activity that is related to Decrease of theta cordance at prefrontal sites after
absolute and relative power values brain perfusion treatment initiation is associated with response
ERPs Amplitude, latency and scalp Assessment of time-locked responses to Several potentials have been found to have a diagnostic or
distribution of averaged, stimulus- external stimuli predictive value
induced EEG waves
LDAEP Difference between N1-P2 amplitude Assessment of the serotonergic function A strong pre-treatment LDAEP predicts favourable
in response to increased stimulus treatment response
intensities
⫹ P50 Positive deflection after approximately Assessment of sensory ‘gating’ and filtering of Increased amplitudes in patients with MDD or bipolar
50 ms incoming information disorder
⫹ P300 Positive deflection after approximately Assessment of attentional aspects and Decreased amplitude and increased latencies are found in
300 ms processing of incoming information MDD
Sleep EEG Neurophysiologically defined sleep Assessment of neurophysiological sleep Disturbed sleep patterns in MDD
stages properties

⫹ REM sleep Latency and density of REM Characterization of REM sleep episodes Increased REM sleep density and decreased REM sleep
episodes latency in MDD
⫹ SWA E.g. DSR Quantification of slow EEG activity during Association between SWA and treatment response
different sleep stages

DSR, delta sleep ratio; ERPs, event-related potentials; FFT, fast Fourier transformation; LDAEP, loudness-dependent auditory evoked potential; LORETA, low resolution electromagnetic
tomography; MDD, major depressive disorder; qEEG, quantitative electroencephalogram; SWA, slow wave activity.
 EEG biomarkers in major depressive disorder 607
et al., 1996; Arns et al., 2012). Within the fast fre- confirmed the asymmetry differences based on intra-
quencies, some evidence exists of increased beta cortical source estimates thus far. However, many
range activity in depressed patients (Knott et al., studies have failed to replicate the findings of alpha
2001; Lieber & Prichep, 1988). asymmetry in MDD (Carvalho et al., 2011; Gold
To overcome often-cited limitations of compari- et al., 2013; Price et al., 2008; Reid et al., 1998; Seg-
sons between topographic EEG study due to differ- rave et al., 2011).
ent references or montages, some studies used By taking a closer look at the often cited Henriques
estimates of intracortical EEG sources such as, and Davidson (1991) data, these researchers used
for example, low resolution brain electromagnetic data from 15 MDD and 13 HC. They reported sig-
tomography (LORETA). This approach has resulted nificant differences in alpha asymmetry between
in the most consistent finding with an altered theta depressive patients and controls on the group level,
activity in frontal areas, most specifically the ACC however, they reported that only two out of 13 HCs
(Jaworska et al., 2012a; Korb et al., 2008; Mientus (15%) deviated significantly from depressive asym-
et al., 2002). metry scores and only one out of 15 depressives (7%)
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

deviated significantly from normal asymmetry scores.

Therefore, there is more overlap between groups
Alpha asymmetry
than there are true differences. This clearly demon-
A large body of literature has been dedicated to lat- strates, along with the above-mentioned failures to
eralized frontal activity in MDD, often referred to as replicate these findings, that this measure has no
frontal asymmetry. Following the assumption that value for diagnostic purposes, which is also acknowl-
increased alpha activity reflects a resting, non-active edged by Davidson (1998).
state, and a decreased alpha activity is associated Furthermore, measures of frontal alpha asymme-
with an increased activity, it is assumed the MDD is try in MDD are only moderately stable over time
characterized by a hyperactive right prefrontal cor- (Debener et al., 2000; Tomarken et al., 1992), have
tex (lower alpha), and a hypoactive left prefrontal low heritability (Anokhin et al., 2006; Smit et al.,
For personal use only.

cortex (higher alpha). In 1983 a group led by David- 2007), and are influenced by differences in cranial
son started publishing their work on frontal alpha and brain parenchymal asymmetries in bone thick-
asymmetry in depression. They reported a relative ness (Myslobodsky et al., 1989) and differences in
hyperactivation of the right frontal cortex, which was EEG montages (Hagemann et al., 1998, 2001; Reid
not found for the parietal cortex (Schaffer et al., et al., 1998), making this a very problematic EEG
1983). Henriques and Davidson (1990) laid a fur- measure to reliably and consistently differentiate
ther foundation for the concept of frontal alpha MDD from HC. Two studies from the same group
asymmetry in depression, where they consider investigated the prognostic value of alpha asymmetry
‘approach’ and ‘withdrawal’ as the essential basis for and found conflicting results (Bruder et al., 2001,
this asymmetry. ‘The approach system facilitates 2008), which makes the prognostic value of this mea-
appetitive behavior and generates certain forms of sure questionable.
positive affect. The withdrawal system facilitates the
withdrawal of an organism from sources of aversive
EEG vigilance
stimulation and generates certain forms of negative
affect’ (Davidson, 1998, page 608). These two sys- In the resting state after closing the eyes a transition
tems have been conceptualized as relatively orthogo- takes place from wakefulness to sleep onset, paral-
nal. They interpreted the decreased left-sided frontal leled by temporal and spatial changes of EEG activity
activation as a deficit in the approach system, and (EEG vigilance stages). These changes of frequency
hence subjects with this condition are more prone over time and space yield important information that
to certain negative affective states and depressive might be relevant for discrimination between MDD
disorders, given a certain level of environmental and healthy controls or even treatment prediction.
stress. On the other hand, they suggested that the Recently Hegerl & Hensch (2012) reinvigorated a
right-sided frontal activation is related to withdrawal- framework that associates a tonically hyperaroused
related emotion and psychopathology such as anxi- central nervous system (CNS) in affective disorders,
ety disorders (Henriques & Davidson, 1990). reflected by a lack of a decline of EEG-based vigi-
In line with this hypothesis, a decreased alpha lance during rest with clinical symptoms of MDD
power at right frontal sites relative to the left side has such as social withdrawal and sensation avoidance
been reported (Chang et al., 2012; Flor-Henry, 1976; (Bente, 1964). Earlier, Ulrich & Fürstenberg (1999)
Henriques & Davidson, 1991; Schaffer et al., 1983). already demonstrated that different subtypes of
Besides the findings from Saletu et al. (2005), who depression could be separated, with the ‘organic sub-
reported increased right frontal activity in postmeno- type’ yielding a hyperstable or rigid EEG vigilance in
pausal depressed women via LORETA, no study comparison to HCs. This was further confirmed by
 608 S. Olbrich & M. Arns
Hegerl et al. (2011) and Olbrich et al. (2012), who stimulus intensities. It has been demonstrated that
demonstrated that patients with MDD exhibited an high basal levels of serotonin in the CNS are related
increased tonic vigilance level quantified via EEG as to a suppression of AEP responsiveness (Hegerl &
compared to HCs. Although more studies are needed Juckel, 1993). Conversely, low serotonin levels are
to confirm the validity of the EEG vigilance measure, reflected in the facilitation of response curves. How-
this approach seems promising since it links clinical ever, most studies failed to find alterations of LDAEP
symptoms of MDD at the behavioural level, underly- in MDD in comparison to HCs (Jaworska et al.,
ing pathomechanisms and electrophysiological find- 2012b; Linka et al., 2007; Park et al., 2010).
ings. Furthermore, from a conceptual level these Another ERP measure is the P300 that is gener-
findings tend to be in line with the often reported ated after presentation of rare or infrequent stimuli
excess alpha which is consistently found in MDD in a so-called oddball paradigm. It is thought to
relative to controls, albeit this EEG vigilance model reflect attentional aspects or processing of incoming
adds a theoretical explanation relating EEG activity information and has been reported to be smaller in
to behaviour; for review also see (Hegerl & Hensch, amplitude in MDD by the majority of studies (e.g.
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

2012). Interestingly, Zoon et al. (2013) found an indi- Blackwood et al., 1987; Diner et al., 1985; Roth
rect linkage of the BDNF Val66Met polymorphism et al., 1981). Bruder et al.(2002) compared the P300
and depression severity via absolute EEG alpha power amplitude between patients with depression only,
in 107 MDD patients. In this study, a decreased EEG anxiety only, a co-morbid group, and a control group,
alpha power during the resting state, possibly related and demonstrated that as compared to the healthy
to altered EEG vigilance stages, was associated with control group the ‘anxiety group’ specifically exhib-
more severe depression ratings while Val66Met poly- ited an increased P300 amplitude, the MDD-only
morphism was related to lower EEG alpha power. group a reduced P300 amplitude, and the MDD
group with co-morbid anxiety showed no differences
in P300 amplitude, demonstrating the need to take
Event-related potential
co-morbidities such as anxiety into account when
For personal use only.

Event-related potentials are averaged responses to assessing P300 amplitudes.

stimuli of various sensory modalities that are recorded Generally, larger reductions in P300 amplitude are
via EEG. The amplitude, latency and scalp distribu- reported in melancholic depression, psychotic
tion of different potential waveforms (Fig. 1) allows depression and depression with suicidal features.
analysis and quantification of the processing of exter- Furthermore, a prolonged P300 latency is often
nal stimuli at a high temporal resolution. found in depression (Bruder et al., 2009; Vandoolae-
The loudness-dependent auditory evoked poten- ghe et al., 1998) but seems not to be influenced by
tial (LDAEP) is a measure reflecting the difference the depressive state, and can hence be seen as a trait
between N1-P2 amplitude in response to increased rather than a state marker (Kalayam & Alexopoulos,
1999; Taylor et al., 2006).
Also, the early potential P50, a measure for sen-
sory gating and early distinction between relevant
and irrelevant stimuli has been found to be altered
in MDD. Baker et al. (1990) reported of impaired
sensory gating in depressive and manic patients, and
Sánchez-Morla et al. (2008) repeated these findings
with increased P50 amplitudes as markers for
impaired suppression of irrelevant stimuli in bipolar
patients. Further, Wang et al. (2009) reported
increases in patients with MDD in comparison to
HCs without a predictive value.
Although no differences of the N400, a potential
associated with semantic processing, have been
revealed between patients with MDD and healthy
subjects in different studies (Deldin et al. 2006;
Klumpp et al. 2010), Ryau et al. (2012) reported
differences between manic patients and patients with
schizophrenia. Findings of other ERP components
from different paradigms, e.g. altered mismatch
Fig. 1. Schematic event-related potential (ERP) wave of an
negativity with decreased amplitudes in MDD
auditory oddball paradigm showing the typical designation of (Chang, Y. et al., 2011; Naismith et al. 2012), further
components. highlight the possibilities of these markers as potential
 EEG biomarkers in major depressive disorder 609
biomarkers in depression, although some studies Further, decreased SWS, decreased sleep effi-
failed to find ERP correlates of behavioural differ- ciency and delayed sleep onset were found to predict
ences between MDD and HCs (Quinn et al., 2012). recurrence of depressive symptoms (Emslie et al.,
2001; Hatzinger et al., 2004; Jindal et al., 2002)
while long REM latency and decreased REM density
EEG connectivity measures
were predictive for the development of depression in
The working brain is not only defined via the location high risk subjects (Rao et al., 2009).
and magnitude of activation clusters but also through While some of these features of sleep EEG have
the interaction of neural activity between different been reported to diminish after treatment with anti-
areas. The EEG provides information about the cou- depressants (Jindal et al., 2003; Kluge et al., 2007;
pling between distinct cortical areas with non-linear Quera Salva et al., 2007), there is evidence that
(phase synchronization) and linear (amplitude) prop- increased REM density and decreased REM latency
erties of the time series that allow analysis of relevant (Friess et al., 2008; Modell et al., 2005; Rao et al.,
network activity in EEG data. Some early studies 2009) are markers of vulnerability in subjects with
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

report differences of one of the first EEG connectiv- high risk for MDD and could thus be considered
ity measures, namely the EEG coherence between ‘trait’ markers.
patients with MDD and HCs (Lieber, 1988;
O’Connor et al., 1979). Since then, different mea-
EEG biomarkers and treatment prediction
sures (e.g. partial directed coherence, Granger cau-
sality, structural synchrony index, phase synchrony In the above overview, QEEG and EEG markers that
index) have emerged that might help to assess alter- can differentiate between MDD and healthy subjects
ation of EEG-based connectivity in MDD. Many have been discussed in more detail. In the following
investigators (Knott et al., 2001; Lee et al., 2011; section we will further explore which measures also
Park et al., 2007; Sun et al., 2008) report of decreased hold prognostic value in predicting treatment out-
EEG coherence measures in MDD. In contrast, other come to various antidepressant treatments.
For personal use only.

research groups found an increased EEG connectiv-

ity in MDD, most consistently in the alpha band
Baseline markers
(Fingelkurts et al., 2007; Jeong et al., 2013; Leuchter
et al., 2012). To clarify the meaning of contrary direc- Quantitative EEG and LORETA. In 1997 Mayberg
tion of alterations (i.e. decrease or increase of con- et al. reported that increased pre-treatment resting
nectivity measures), critical discussions about the metabolism of the rostral anterior cingulate (BA
impact of EEG amplitude and volume conduction 24a/b) predicted favourable treatment response to
on the different connectivity measures are needed. antidepressants. Since then, this has sparked a huge
Then EEG connectivity-based measures could prob- research interest into the link between the ACC and
ably serve as reliable and valid biomarkers. treatment response in depression, and to date this is
the most investigated finding in treatment prediction
in depression. In order to integrate all these results
Sleep research
recently a meta-analysis was performed (Pizzagalli,
MDD is often accompanied by sleep disturbances 2011), including 23 studies. A total of 19 studies
(Benca et al., 1992). Therefore it is not surprising that reported that responders to antidepressant treatment
quite robust findings of EEG research in MDD stem (medication, electroconvulsive therapy (ECT),
from sleep research. Consistent results from several repetitive transcranial magnetic stimulation (rTMS)
decades of sleep research report a disturbed sleep or sleep deprivation (SD)) demonstrated increased
architecture comprising an increased rapid eye move- ACC activity pre-treatment, whereas the remaining
ment (REM) density (Goetz et al., 1991; Lauer et al., four studies found the opposite. The overall effect size
1991), a shortened REM sleep latency (Reynolds (ES) was large (ES ⫽ 0.918). The relationship
et al., 1985; Rotenberg et al., 2002) and altered slow between increased ACC activity and favourable anti-
wave sleep (SWS) in MDD (Hoffmann et al., 2000; depressant response was found consistently across
Lopes et al., 2007), though, others found no REM treatments (selective serotonin reuptake inhibitors
sleep increases in a study of 67 male MDD patients (SSRI), tricyclic antidepressants (TCA), ketamine,
(Hubain et al., 2006). However, the first non-REM rTMS and SD), imaging modalities (EEG-LORETA,
sleep phase especially seems to have a diagnostic fMRI and SPECT) and did not depend on medica-
value: Armitage et al. (2001) reported that amplitude tion status at baseline. No clear relationship between
differences of delta power within the first REM period activity in the anterior cingulate and specific neu-
differentiated between MDD and HCs, and Hoff- rotransmitter systems has been repor ted (Mulert
mann et al. (2000) showed that slow wave activity et al., 2007) and treatment-resistant depressive
was lower during the first REM period in MDD. patients have also been shown to respond to deep
 610 S. Olbrich & M. Arns
brain stimulation of ACC areas (see Hamani et al., as a function of stimulus intensity that indicates low
2011 for a review) suggesting that ACC activity serotonergic function has been demonstrated to pre-
reflects a reliable biomarker for antidepressant treat- dict a larger decrease of depressive symptoms after
ment response in general. treatment with SSRIs (Gallinat et al., 2000; Juckel
Increased theta in the ACC imaged with LORETA et al., 2007; Lee et al., 2005; Mulert et al., 2002). In
has been shown to reflect increased metabolism in contrast to the hypothesis of a normalized seroton-
the ACC (Pizzagalli et al., 2003) and is thus a reliable ergic transmission after treatment, a retest after 4
predictor for a favourable treatment outcome. Fur- weeks did not show a change of the LDAEP (Gallinat
thermore, ACC theta activity also often shows up as et al., 2000).
frontal midline theta, thus excess theta at electrode Also, a reduced P300 amplitude has been reported
sites Fz or FCz has been found to be associated with to be associated with a poor treatment outcome to
a favourable treatment outcome (Spronk et al., antidepressant medication (Bruder et al., 1995,
2011). This frontal midline theta should not be con- 2001) and ECT (Ancy et al., 1996; Gangadhar
fused with wide-spread frontal theta, which is more et al., 1993). Furthermore, a prolonged P300 latency
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

often a reflection of drowsiness, or as discussed has been found to be associated with a poor treat-
above, a sign of low vigilance. This widespread fron- ment outcome (I ınta et al. 2012; Kalayam & Alex-
tal excess theta has often been found to be associated opoulos, 1999; Vandoolaeghe et al., 1998). Several
with non-response to antidepressant treatments, also authors have suggested that these non-responders to
see below for further details. antidepressant medication can be regarded as a sub-
In QEEG research, various pre-treatment differ- group with ‘pre-frontal dysfunction’ (Dunkin et al.,
ences in EEG measures have been reported to be 2000; Kalayam & Alexopoulos, 1999; Vandoolaeghe
associated with improved antidepressant treatment et al., 1998).
outcomes. Biomarkers associated with poor antide- Taylor et al. (2006) reported that SSRI non-
pressant response which have at least been replicated responders were characterized by more psychomotor
once include: slowing as compared to responders, and interpreted
For personal use only.

this to be related to reduced dopaminergic neu-

1. Decreased parieto-occipital alpha power (SSRI:
rotransmission. In this respect the P300 latency
Bruder et al., 2008; TCA: Ulrich et al., 1984), also
might also be seen as an electrophysiological mea-
using current density maps (Tenke et al., 2011)
sure of psychomotor speed, hence suggesting that a
and decreased frontal alpha power (Suffin &
combination of a slow P300 latency and a slowed
Emory, 1995). In line with the results summarized
reaction time points to an underlying dopaminergic
in the previous section on EEG findings which
rather than serotonergic problem. This was under-
can discriminate between MDD and HCs, this
lined by a reported association between P300 ampli-
finding can thus be considered an atypical MDD
tude and latencies and dopamine D2/D3 receptor
group, given the consistent finding of excess alpha
status (Pogarell et al., 2011) in MDD, and might
in MDD. Hence it is interesting to note that MDD
explain why this subgroup of patients is less respon-
patients characterized by decreased alpha do not
sive to antidepressant medication. On the other hand,
respond well to antidepressant treatments.
a recent study failed to find a relationship between
2. Increased slow EEG power at baseline. Increased
the P300 and dopaminergic genes (Spronk et al.,
theta (rTMS: Arns et al., 2012; TCA: Knott
2013) suggesting the association with dopamine and
et al., 1996), increased relative theta (SSRI &
the P300 is not that straightforward.
serotonin–norepinephrine reuptake inhibitors
Other ERP measures such as increased amplitude
(SNRI): Iosifescu et al., 2009) and increased
of somatosensory-evoked potentials after infusion of
delta power (SSRI: Knott et al., 2000; TCA-
ketamine in responders (Cornwell et al., 2012) or an
trend: Knott et al., 1996). However, Cook et al.
increased N100 amplitude (Danos et al. 1994) have
(1999) found no differences in theta for respond-
also been reported to have some value in the predic-
ers and non-responders to fluoxetine.
tion of treatment.
3. A slow individual alpha peak frequency (iAPF)
for antidepressant medication (Ulrich et al.,
1984) and rTMS treatment (Arns et al., 2010, Sleep EEG
2012; Conca et al., 2000).
In sleep research, the slow wave activity (SWA) seems
to play an important role for treatment prediction.
Luthringer et al. (1995) reported increased relative
Event-related potentials
delta power in sleep EEG recordings in responders
Also from the field of evoked potentials, some prom- to antidepressant treatment, although others failed to
ising measures that correlate with treatment response find power differences between responders and non-
have been published. A strong pre-treatment LDAEP responders to psychotherapy and antidepressant
 EEG biomarkers in major depressive disorder 611
treatment (Buysse et al., 2001) or sleep deprivation treatment response has not been addressed so far
(Nissen et al., 2001). Still, the latter reported of (Kuo & Tsai, 2010).
decreasing SWA in responders for consecutive non-
REM episodes of one night, expressed in a high delta
Discussion
sleep ratio (DSR), a finding that could not be repeated
by Argyropulos et al. (2009). The amount of selective From the first observations of altered EEG related
suppression of frontal SWA through acoustic stimuli to affective capacity in 1938 (Lemere, 1936) to more
was also found to correlate with improvements from controlled research in MDD now, is a long way from
baseline depression severity (Landsness et al., 2011). the mere ‘qualitative’ description of altered alpha
In contrast, a low DSR predicting treatment response activity to complex ‘quantitative’ measures such as
was found recently by Duncan et al. (2013) in oth- cordance and P300. Steady research has tried to
erwise treatment-resistant patients with MDD after sharpen the tools for discriminating patients with
infusion of ketamine. At first glance these heteroge- MDD from healthy controls and to identify possible
neous findings probably reflect the impact of treat- EEG-based markers for treatment prediction to
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

ment methods on the predictive value of EEG shorten treatment paths and improve efficacy of
biomarkers. Besides classical sleep EEG parameters, treatment.
a decreased coherence within the beta, delta and In the field of MDD, especially, differences in the
theta band in sleep EEG also predicted non-response EEG alpha and theta frequency range during rest
in adolescents and the occurrence of depressive epi- have been reported consistently, and both these mea-
sodes (Morehouse et al., 2002). sures have also been differentially associated with
treatment outcome. Still, the interpretation of these
measures remains vague. On one hand EEG alpha
Treatment emergent markers
power has been associated with an idling function
EEG cordance. Within the last decade it was found that reflects disengagement of cortical areas from
that not only baseline measurements might help to active information processing and hence increased
For personal use only.

predict favourable treatment outcome but also the alpha power at occipital or frontal areas has been
assessment of changes of EEG activity after treat- interpreted as a subvigil brain state in MDD. On the
ment onset. For the EEG cordance, a complex mea- other hand, increased EEG alpha power during rest
sure that reflects the association between relative has been associated with increased EEG vigilance in
and absolute EEG power and has been associated MDD and altered decline towards decreased vigi-
with brain perfusion (Leuchter et al., 1999), it was lance states with elevated theta and delta activity.
reported by Leuchter et al. (1997) and Rabinoff Also, the results from sleep research show diminished
et al. (2011) that 48 h to 1 week after treatment drops toward deep sleep stages with decreased SWA
initiation, patients with MDD who later showed a during sleep in MDD and altered REM sleep pro-
clinical response yielded a decrease of theta cor- files. Thus, EEG-based biomarkers of MDD in gen-
dance at prefrontal sites in comparison to baseline eral might reflect a rigid and less flexible CNS (to
(also Bares et al., 2007, 2008; Cook et al., 2002, externally and internally forced changes of brain
2005; Leuchter et al., 2002). This was not the case function) that leads to impaired behavioural adap-
in non-responders or patients treated with placebo tion of the whole organism to the requirements of its
(Leuchter et al., 2002). Hunter et al. (2010) also environment in major depression.
report a decrease of midline and right frontal cor- However, discrepancies of findings from all EEG-
dance after 48 h of antidepressant treatment in based measures remain and surely reflect different
patients with suicidal ideations. underlying mechanisms and subgroups that are not
represented within diagnostic systems such as
DSM-IV and ICD-10. Still, some of the divergent
Antidepressant treatment response index
findings could possibly be resolved if changes of neu-
A further marker of treatment response worth men- rophysiologically assessed brain function would be
tioning is the antidepressant treatment response interpreted in a more generalized manner rather than
(ATR) index, a measure that combines relative and trying to trace them down to specific cognitive func-
absolute alpha power from different montages. tions or barely interpretable terms. Different inter-
Developed by the same research group as the cor- pretations of findings (e.g. alpha activity as a marker
dance measure, it has been found to be predictive for generalized hyperstable vigilance or as a sign of
for remission or response in a large trial with n ⫽ 376 less vigilant focal brain areas), reveal the lack of com-
(Leuchter et al., 2009). The prediction of response monly accepted models about the meaning of EEG-
was later confirmed in a small placebo controlled based markers. Thus, integrated frameworks such as
trial (Hunter et al., 2011). The question whether the EEG vigilance model that combine neurophysi-
ATR or cordance performs better in predicting ological findings with clinical symptoms are urgently
 612 S. Olbrich & M. Arns
needed to refine and better understand widely measures might be of particular use for guiding the
accepted biomarkers. By showing an indirect linkage clinical decision for a special antidepressant agent.
between the BDNF Val66Met polymorphism and From the clinical point of view, EEG-based bio-
symptom severity of MDD that is mediated through markers for differentiation between patients and
(possibly vigilance-dependent) resting state EEG healthy subjects are only of limited value because the
alpha power, Zoon et al. (2013) also substantially diagnosis is derived from clinical criteria. Although
contribute to a holistic understanding of neurophys- these biomarkers can help to understand the multiple
iological biomarkers in MDD. patho-mechanisms that lead to MDD, the main focus
The first steps towards an integration of concepts of EEG biomarker research has shifted to the progno-
also requires taking into account different aspects of sis of the course of illness during treatment. And
brain activity. Comparisons with findings from other indeed, remarkable progress has been made with sev-
neuroimaging modalities are necessary and can be eral biomarkers that have shown a consistently predic-
achieved via studies that do not rely on montage and tive value. As a baseline marker, especially rACC theta,
reference-dependent measures, but instead use cur- the LDAEP, the iAPF, the P300 and frontal theta
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

rent density maps or estimated sources of scalp activity have been found to predict treatment response
EEG. For example, the increased frontal midline to specific agents such as SSRIs or SNRIs, while cor-
theta activity in MDD has been traced to the rostral dance measures as well as the ATR show notable suc-
ACC, reflective of increased metabolic activity, and cess as treatment-emergent biomarkers and predicting
confirmed with other imaging techniques such as treatment response in the early course of treatment.
fMRI, PET and SPECT imaging as reviewed in a However, several issues, especially standardized
meta-analysis from Pizzagalli (2011). This replica- procedures of recording condition, pre-processing
tion from other imaging modalities enhances the and analysis of data should be on the agenda on the
interpretation of findings and strengthens the gen- way towards valid and clinically useful biomarkers for
eration of hypothesis about possible biomarkers. treatment prediction to prevent this field of research
Further, region of interest (ROI) based analysis of from the blunting incomparability of studies:
For personal use only.

EEG source estimates could increase the statistical

power for research on EEG biomarkers. More work • Length of recording and length of used EEG
is also needed to add information of cortical neuro- epochs for analysis should be unified
transmitters, e.g. glutamate and gamma-aminobutyric • The order in which EEG is collected (i.e. is the
acid, and excitation – inhibition balance, measured eyes open EEG or the eyes closed EEG recorded
via MRI spectroscopy or paired-pulse TMS proto- as the first task, or after 30–60 min of neuropsy-
cols (Tremblay et al., 2013) to models of altered chological tests?)
EEG-profiles in MDD. The integration of concepts • Used channels and channel numbers, and espe-
might further benefit from biomarkers of neuronal cially how ROIs are defined (voxel-based, sphere,
interaction such as EEG-based coherence and rela- etc.) should be unified, especially for the use of
ted markers. These measures provide insight into source estimation techniques due to the depen-
the coupling of distant brain areas and allow the dency of these tools to channel distribution
analysis of altered network function. Especially • The condition (resting, trying or not trying to fall
within fMRI research, connectivity analysis is a rap- asleep, interfere or not interfere when vigilance
idly growing field. It should be in the focus of EEG declines) and the environment (light, noise, tem-
research to identify neurophysiological correlates of perature) should be kept constant
altered fMRI brain connectivity because the EEG • Pre-processing procedures (e.g. correction or
would allow the translation of these findings into removal of artefact segments) and the choice of
clinical relevant biomarkers. segments (whole EEG record with artefact mini-
ERP-based approaches for identification of bio- mization or analysis of selected artefact-free
markers might also be of value due to the reported epochs) should be clarified
association of different potentials to neurotransmit- • Measures used should be similar or at least com-
ter function. Although the LDAEP has failed to dis- parable throughout different studies
tinguish between patients and controls in general, its • Due to the impact of circadian rhythms on EEG
correlation to the serotonergic function might serve profiles (e.g. via vigilance changes), time of
for the separation of biologically different subgroups recording should be standardized and further be
(Mulert et al., 2007). Further more, the P300 is dis- comparable to the working schedule in clinical
cussed to reflect, at least in parts, dopaminergic EEG laboratories, e.g. during the morning or
transmission (Pogarell et al., 2011), although findings early afternoon
are inconsistent (Spronk et al., 2013). However, It should be noticed that the International Phar-
since antidepressant medications interfere with brain maco EEG Society (IPEG) recently released guide-
function at different neurotransmitter receptors, ERP lines (Jobert et al., 2012; 2013) that help to make
 EEG biomarkers in major depressive disorder 613
obtained results from different recording sites and Argyropoulos, S.V., Hicks, J.A., Nash, J.R., Bell, C.J., Rich, A.S.,
study centres comparable; for ERP studies the Inter- Nutt, D.J. & Wilson, S. (2009). Redistribution of slow wave
activity of sleep during pharmacological treatment of depres-
national Federation of Clinical Neurophysiology has sion with paroxetine but not with nefazodone. Journal of Sleep
released a guideline for event-related potentials in Research, 18, 342–348.
clinical research (Duncan et al. 2009). The value of Armitage, R., Emslie, G.J., Hoffmann, R.F., Rintelmann, J. &
upcoming works on EEG-based biomarkers in MDD Rush, A.J. (2001). Delta sleep EEG in depressed adolescent
should also be judged by the comparability with females and healthy controls. Journal of Affective Disorders, 63,
139–148.
other studies and by the accordance with standard- Arns, M., Drinkenburg, W.H., Fitzgerald, P.B. & Kenemans, J.L.
ized procedures. (2012). Neurophysiological predictors of non-response to
Besides the standardization of EEG biomarker rTMS in depression. Brain Stimulation 5, 569–576.
research, there are other aspects that could further Arns, M., Spronk, D. & Fitzgerald, P.B. (2010). Potential differ-
enhance the value and validity of EEG biomarker ential effects of 9 hz rtms and 10 hz rtms in the treatment of
depression. Brain Stimulation, 3, 124–126.
research. At first, most of the studies up to now Atkinson, A.J., Colburn, W.A., DeGruttola, V.G., DeMets, D.L.,
included only a limited number of subjects and there Downing, G.J., Hoth, D.F., … Zeger, S.L. (2001). Biomarkers
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

is a clear lack of multi-centre studies in EEG research, and surrogate endpoints: Preferred definitions and conceptual
a fact that is a clear deficiency of EEG research in framework. Clinical Pharmacology & Therapeutics, 69,
comparison to biomarker research in other fields of 89–95.
Baker, N.J., Staunton, M., Adler, L.E., Gerhardt, G.A.,
medicine. It should be a future goal to enrol large, Drebing, C., Waldo, M., … Freedman, R. (1990). Sensory gat-
hypothesis-driven cohort studies to test the validity ing deficits in psychiatric inpatients: Relation to catecholamine
of already existing biomarkers, such as the multi- metabolites in different diagnostic groups. Biological Psychiatry,
centre international Study to Predict Optimized 27, 519–528.
Treatment (iSPOT) (Williams et al., 2011). These Bares, M., Brunovsky, M., Kopecek, M., Novak, T., Stopkova, P.,
Kozeny, J., … Höschl, C. (2008). Early reduction in prefrontal
studies should not necessarily focus on one measure theta QEEG cordance value predicts response to venlafaxine
but should compare combinations of markers for treatment in patients with resistant depressive disorder. Euro-
their discriminative and predictive power, i.e. inte- pean Psychiatry, 23, 350–355.
For personal use only.

grative neuroscience (Kuo & Tsai, 2010; Spronk Bares, M., Brunovsky, M., Kopecek, M., Stopkova, P., Novak, T.,
et al., 2011). The search for predictive biomarkers Kozeny, J. & Höschl, C. (2007). Changes in QEEG prefrontal
cordance as a predictor of response to antidepressants in
could also benefit from approaches that incorporate patients with treatment resistant depressive disorder: A pilot
combinations of EEG-based and non-EEG markers study. Journal of Psychiatric Research, 41, 319–325.
as demonstrated by Spronk et al. (2011). Further- Begić, D., Popović-Knapić, V., Grubišin, J., Kosanović-Rajačić, B.,
more, used measures should not be too complex Filipčić, I., Telarović, I. & Jakovljević, M. (2011). Quantitative
since simple description of the underlying mecha- electroencephalography in schizophrenia and depression. Psy-
chiatria Danubina, 23, 355–362.
nisms that lead to neurophysiologically assessable Benca, R.M., Obermeyer, W.H., Thisted, R.A. & Gillin, J.C. (1992).
biomarkers would greatly enhance its acceptance by Sleep and psychiatric disorders. A meta-analysis. Archives of
clinicians and help to justify the usage of these bio- General Psychiatry, 49, 651–668; discussion 669–670.
markers for therapeutic interventions in patients. Bente, D. (1964). Vigilanz, dissoziative Vigilanzverschiebung und
Insuffizienz des Vigilitätstonus. [Vigilance, dissociative shift
of vigilance and insufficiency of tonic vigilance]. Kranz, H.,
Declaration of interest: Sebastian Olbrich reports Heinrich, K. (editors) In Begleitwirkungen und Mißerfolge der
no conflict of interest. Martijn Arns reports research psychiatrischen Pharmakotherapie (pp. 13–28). Stuttgart: Thieme.
grants and options from Brain Resource Ltd. Berger, H. (1933). Über das Elektrenkephalogramm des Menschen.
[On the encephalogram of man]. Archiv für Psychiatrie und Ner-
(Sydney, Australia) acted as a paid consultant for
venkrankheiten, 99, 555–574.
United BioSource Corporation (UBC) and Bracket Blackwood, D.H., Whalley, L.J., Christie, J.E., Blackburn, I.M., St
and has been an author on 3 patent applications Clair, D.M & McInnes, A. (1987). Changes in auditory P3
related to EEG and psychophysiology but does not event-related potential in schizophrenia and depression. British
own these patents nor has any future financial gains Journal of Psychiatry, 150, 154–160.
Bruder, G.E., Kayser, J & Tenke, C.E. (2009). Event-related brain
from these patents and these have no relation to the
potentials in depression: Clinical, cognitive and neurophysio-
materials presented. The authors alone are respon- logic implications. In S.J. Luck & E.S. Kappenman (Eds),
sible for the content and writing of the paper. The Oxford Handbook of Event-Related Potential Components
(pp. 563–592). New York: Oxford University Press.
Bruder, G.E., Kayser, J., Tenke, C.E., Leite, P., Schneier, F.R.,
Stewart, J.W. & Quitkin, F.M. (2002). Cognitive ERPs in
References
depressive and anxiety disorders during tonal and phonetic
Ancy, J., Gangadhar, B.N. & Janakiramaiah, N. (1996). ‘Normal’ oddball tasks. Clinical Electroencephalography, 33, 119–124.
P300 amplitude predicts rapid response to ECT in melancho- Bruder, G.E., Sedoruk, J.P., Stewart, J.W., McGrath, P.J.,
lia. Journal of Affective Disorders, 41, 211–215. Quitkin, F.M. & Tenke, C.E. (2008). Electroencephalographic
Anokhin, A.P., Heath, A.C. & Myers, E. (2006). Genetic and envi- alpha measures predict therapeutic response to a selective sero-
ronmental influences on frontal EEG asymmetry: A twin study. tonin reuptake inhibitor antidepressant: pre- and post-treatment
Biological Psychology, 71, 289–295. findings. Biological Psychiatry, 63, 1171–1177.
 614 S. Olbrich & M. Arns
Bruder, G.E., Stewart, J.W., Tenke, C.E., McGrath, P.J., Leite, P., Deldin, P., Keller, J., Casas, B.R., Best, J., Gergen, J. &
Bhattacharya, N. & Quitkin, F.M. (2001). Electroencephalo- Miller, G.A. (2006). Normal N400 in mood disorders.
graphic and perceptual asymmetry differences between respond- Biological psychology, 71, 74–79.
ers and nonresponders to an SSRI antidepressant. Biological Diner, B.C., Holcomb, P.J. & Dykman, R.A. (1985). P300 in
Psychiatry, 49, 416–425. major depressive disorder. Psychiatry Research, 15, 175–184.
Bruder, G.E., Tenke, C.E., Stewart, J.W., Towey, J.P., Leite, P., Duncan, A.C., Barry, R.J., Connolly, J.F., Fischer, C.,
Voglmaier, M. & Quitkin, F.M. (1995). Brain event-related Michie, P.T., Näätänen, R., … Van Petten, C. (2009). Event-
potentials to complex tones in depressed patients: Relations to related potentials in clinical research: Guidelines for eliciting,
perceptual asymmetry and clinical features. Psychophysiology, recording, and quantifying mismatch negativity, P300, and
32, 373–381. N400. Clinical Neurophysiology, 120, 1883–1908.
Buysse, D.J., Hall, M., Begley, A., Cherry, C.R., Houck, P.R., Duncan, W.C. Jr, Selter, J., Brutsche, N., Sarasso, S. &
Land, S., … Frank, E. (2001). Sleep and treatment response in Zarate, C.A. Jr. (2013). Baseline delta sleep ratio predicts acute
depression: New findings using power spectral analysis. ketamine mood response in major depressive disorder. Journal
Psychiatry Research, 103, 51–67. of Affective Disorders, 145, 115–119.
Carvalho, A., Moraes, H., Silveira, H., Ribeiro, P., Dunkin, J.J., Leuchter, A.F., Cook, I.A., Kasl-Godley, J.E.,
Piedade, R.A.M., Deslandes, A.C., … Versiani, M. (2011). EEG Abrams, M. & Rosenberg-Thompson, S. (2000). Executive dys-
frontal asymmetry in the depressed and remitted elderly: is function predicts nonresponse to fluoxetine in major depres-
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

it related to the trait or to the state of depression? Journal of sion. Journal of Affective Disorders, 60, 13–23.
Affective Disorders, 129, 143–148. Emslie, G.J., Armitage, R., Weinberg, W.A., Rush, A.J., Mayes, T.L.
Chang, C.C., Yu, S.C., McQuoid, D.R., Messer, D.F., & Hoffmann, R.F. (2001). Sleep polysomnography as a predic-
Taylor, W.D., Singh, K., … Payne, M.E. (2011). Reduction of tor of recurrence in children and adolescents with major depres-
dorsolateral prefrontal cortex gray matter in late-life depression. sive disorder. International Journal of Neuropsychopharmacology,
Psychiatry Research, 193, 1–6. 4, 159–168.
Chang, J.S., Yoo, C.S., Yi, S.H., Her, J.Y., Choi, H.M., Ha, T.H., Fingelkurts, A.A., Fingelkurts, A.A., Rytsälä, H., Suominen, K.,
… Ha, K. (2012). An integrative assessment of the psycho- Isometsä, E. & Kähkönen, S. (2007). Impaired functional con-
physiologic alterations in young women with recurrent major nectivity at EEG alpha and theta frequency bands in major
depressive disorder. Psychosomatic Medicine, 74, 495–500. depression. Human Brain Mapping, 28(3), 247–261.
Chang, Y., Xu, J., Shi, N., Pang, X., Zhang, B. & Cai, Z. (2011). Flor-Henry, P. (1976). Lateralized temporal-limbic dysfunction
Dysfunction of preattentive visual information processing and psychopathology. Annals of the NewYork Academy of Sciences,
among patients with major depressive disorder. Biological 280, 777–795.
For personal use only.

Psychiatry, 69, 742–747. Flor-Henry, P., Koles, Z.J., Howarth, B.C. & Burton, L. (1979).
Conca, A., Swoboda, E., König, P., Koppi, S., Beraus, W., Neurophysiological studies of schizophrenia, mania and depres-
Künz, A., . Weiss, P. (2000). Clinical impacts of single transcra- sion. In J. Gruzelier & P. Flor-Henry, (Eds), Hemisphere Asym-
nial magnetic stimulation (STMS) as an add-on therapy in metries of Function in Psychopathology (pp. 189–222). Amsterdam:
severely depressed patients under SSRI treatment. Human Psy- Elsevier.
chopharmacology, 15, 429–438. Fox, M.D., Liu, H. & Pascual-Leone, A. (2012). Identification of
Cook, I.A., Leuchter, A.F., Morgan, M.L., Stubbeman, W., reproducible individualized targets for treatment of depression
Siegman, B. & Abrams, M. (2005). Changes in prefrontal activ- with TMS based on intrinsic connectivity. NeuroImage, 66C,
ity characterize clinical response in SSRI nonresponders: 151–160.
A pilot study. Journal of Psychiatric Research, 39, 461–466. Friess, E., Modell, S., Brunner, H., Tagaya, H., Lauer, C. J.,
Cook, I.A., Leuchter, A.F., Morgan, M., Witte, E., Holsboer, F. & Ising, M. (2008). The Munich vulnerability
Stubbeman, W.F., Abrams, M., … Uijtdehaage, S.H.J. (2002). study on affective disorders: Microstructure of sleep in high-risk
Early changes in prefrontal activity characterize clinical respond- subjects. European Archives of Psychiatry and Clinical Neuro-
ers to antidepressants. Neuropsychopharmacology, 27, 120–131. science, 258, 285–291.
Cook, I.A., Leuchter, A.F., Witte, E., Abrams, M., Uijtdehaage, Gallinat, J., Bottlender, R., Juckel, G., Munke-Puchner, A.,
S.H., Stubbeman, W., . Dunkin, J.J. (1999). Neurophysiologic Stotz, G., Kuss, H.J., … Hegerl, U. (2000). The loudness
predictors of treatment response to fluoxetine in major depres- dependency of the auditory evoked N1/P2-component as a pre-
sion. Psychiatry Research, 85, 263–273. dictor of the acute SSRI response in depression. Psychopharma-
Cornwell, B.R., Salvadore, G., Furey, M., Marquardt, C.A., cology, 148, 404–411.
Brutsche, N.E., Grillon, C. & Zarate, C.A., Jr. (2012). Synaptic Gangadhar, B.N., Ancy, J., Janakiramaiah, N. & Umapathy, C.
potentiation is critical for rapid antidepressant response to (1993). P300 amplitude in non-bipolar, melancholic depres-
ketamine in treatment-resistant major depression. Biological sion. Journal of Affective Disorders, 28, 57–60.
Psychiatry, 72, 555–561. Gatt, J.M., Kuan, S.A., Dobson-Stone, C., Paul, R.H., Joffe, R.T.,
Danos, P., Kasper, S., Scholl, H.P., Kaiser, J., Ruhrmann, S., Kemp, A.H., … Williams, L.M. (2008). Association between
Höflich, G. & Möller, H.J. (1994). Clinical response to sleep BDNF Val66Met polymorphism and trait depression is medi-
deprivation and auditory-evoked potentials – Preliminary ated via resting EEG alpha band activity. Biological Psychology,
results. Pharmacopsychiatry, 27, 70–71. 79, 275–284.
Davidson, R.J. (1998). Anterior electrophysiological asymmetries, Goetz, R.R., Puig-Antich, J., Dahl, R.E., Ryan, N.D., Asnis, G.M.,
emotion, and depression: Conceptual and methodological Rabinovich, H. & Nelson, B. (1991). EEG sleep of young adults
conundrums. Psychophysiology, 35, 607–614. with major depression: A controlled study. Journal of Affective
Debener, S., Beauducel, A., Nessler, D., Brocke, B., Heilemann, H. Disorders, 22, 91–100.
& Kayser, J. (2000). Is resting anterior EEG alpha asymmetry a Gold, C., Fachner, J. & Erkkilä, J. (2013). Validity and reliability
trait marker for depression? Findings for healthy adults and of electroencephalographic frontal alpha asymmetry and frontal
clinically depressed patients. Neuropsychobiology, 41, 31–37. midline theta as biomarkers for depression. Scandinavian Jour-
De Gennaro, L., Marzano, C., Fratello, F., Moroni, F., nal of Psychology, 54, 118–126.
Pellicciari, M. C., Ferlazzo, F., … Rossini, P.M. (2008). Grin-Yatsenko, V.A., Baas, I., Ponomarev, V.A. & Kropotov, J.D.
The electroencephalographic fingerprint of sleep is genetically (2009). EEG power spectra at early stages of depressive disor-
determined: A twin study. Annals of Neurology, 64, 455–460. ders. Journal of Clinical Neurophysiology, 26, 401–406.
 EEG biomarkers in major depressive disorder 615
Grin-Yatsenko, V.A., Baas, I., Ponomarev, V.A. & Kropotov, J.D. Jaworska, N., Blier, P., Fusee, W. & Knott, V. (2012a). α
(2010). Independent component approach to the analysis of Power, α asymmetry and anterior cingulate cortex activity in
EEG recordings at early stages of depressive disorders. Clinical depressed males and females. Journal of Psychiatric Research, 46,
Neurophysiology, 121, 281–289. 1483–1491.
Hagemann, D., Naumann, E., Becker, G., Maier, S. & Bartussek, Jaworska, N., Blier, P., Fusee, W. & Knott, V. (2012b). Scalp- and
D. (1998). Frontal brain asymmetry and affective style: A con- sLORETA-derived loudness dependence of auditory evoked
ceptual replication. Psychophysiology, 35, 372–388. potentials (LDAEPs) in unmedicated depressed males and
Hagemann, D., Naumann, E. & Thayer, J.F. (2001). The quest for females and healthy controls. Clinical Neurophysiology, 123,
the EEG reference revisited: A glance from brain asymmetry 1769–1778.
research. Psychophysiology, 38, 847–857. Jeong, H.G., Ko, Y.H., Han, C., Kim, Y.K. & Joe, S.-H. (2013).
Hamani, C., Mayberg, H., Stone, S., Laxton, A., Haber, S. & Distinguishing quantitative electroencephalogram findings
Lozano, A.M. (2011). The subcallosal cingulate gyrus in between adjustment disorder and major depressive disorder.
the context of major depression. Biological Psychiatry, 69, Psychiatry Investigation, 10, 62–68.
301–308. Jindal, R.D., Friedman, E.S., Berman, S.R., Fasiczka, A.L.,
Hatzinger, M., Hemmeter, U.M., Brand, S., Ising, M. & Howland, R.H. & Thase, M.E. (2003). Effects of sertraline on
Holsboer-Trachsler, E. (2004). Electroencephalographic sleep sleep architecture in patients with depression. Journal of Clinical
profiles in treatment course and long-term outcome of major Psychopharmacology, 23, 540–548.
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

depression: Association with DEX/CRH-test response. Journal Jindal, R.D., Thase, M.E., Fasiczka, A.L., Friedman, E.S.,
of Psychiatric Research, 38, 453–465. Buysse, D.J., Frank, E. & Kupfer, D.J. (2002). Electroencepha-
Hegerl, U. & Hensch, T. (2012). The vigilance regulation model lographic sleep profiles in single-episode and recurrent unipolar
of affective disorders and ADHD. Neuroscience and Biobehavio- forms of major depression: II. Comparison during remission.
ral Reviews. doi:10.1016/j.neubiorev.2012.10.008 Biological Psychiatry, 51, 230–236.
Hegerl, U. & Juckel, G. (1993). Intensity dependence of auditory Jobert, M., Wilson, F.J., Roth, T., Ruigt, G.S.F., Anderer, P. &
evoked potentials as an indicator of central serotonergic Drinkenburg, W.H.I.M. (2013). Guidelines for the recording
neurotransmission: A new hypothesis. Biological Psychiatry, 33, and evaluation of pharmaco-sleep studies in man: The Interna-
173–187. tional Pharmaco-EEG Society (IPEG). Neuropsychobiology, 67,
Hegerl, U., Wilk, K., Olbrich, S., Schoenknecht, P. & Sander, C. 127–167.
(2011). Hyperstable regulation of vigilance in patients with Jobert, M., Wilson, F.J., Ruigt, G.S.F., Brunovsky, M.,
major depressive disorder. World Journal of Biological Psychiatry, Prichep, L.S. & Drinkenburg, W.H.I.M. (2012). Guidelines for
13(6), 436–446. the recording and evaluation of pharmaco-EEG data in man:
For personal use only.

Henriques, J.B. & Davidson, R.J. (1990). Regional brain electrical The International Pharmaco-EEG Society (IPEG). Neuropsy-
asymmetries discriminate between previously depressed and chobiology, 66, 201–220.
healthy control subjects. Journal of Abnormal Psychology, 99, John, E.R., Prichep, L.S., Fridman, J. & Easton, P. (1988).
22–31. Neurometrics: Computer-assisted differential diagnosis of
Henriques, J.B. & Davidson, R.J. (1991). Left frontal hypoacti- brain dysfunctions. Science (New York), 239, 162–169.
vation in depression. Journal of Abnormal Psychology, 100, Juckel, G., Pogarell, O., Augustin, H., Mulert, C., Müller-
535–545. Siecheneder, F., Frodl, T., … Hegerl, U. (2007). Differential
Hoffmann, R., Hendrickse, W., Rush, A.J. & Armitage, R. (2000). prediction of first clinical response to serotonergic and
Slow-wave activity during non-REM sleep in men with schizo- noradrenergic antidepressants using the loudness dependence
phrenia and major depressive disorders. Psychiatry Research, 95, of auditory evoked potentials in patients with major depressive
215–225. disorder. Journal of Clinical Psychiatry, 68, 1206–1212.
Hubain, P., Le Bon, O., Vandenhende, F., Van Wijnendaele, R. & Kalayam, B. & Alexopoulos, G.S. (1999). Prefrontal dysfunction
Linkowski, P. (2006). Major depression in males: Effects of age, and treatment response in geriatric depression. Archives of
severity and adaptation on sleep variables. Psychiatry Research, General Psychiatry, 56, 713–718.
145, 169–177. Kluge, M., Schüssler, P. & Steiger, A. (2007). Duloxetine increases
Hunter, A.M., Cook, I.A., Greenwald, S.D., Tran, M.L., stage 3 sleep and suppresses rapid eye movement (REM) sleep
Miyamoto, K.N. & Leuchter, A.F. (2011). The antidepressant in patients with major depression. European Neuropsychophar-
treatment response index and treatment outcomes in a placebo- macology, 17, 527–531. doi:10.1016/j.euroneuro.2007.01.006
controlled trial of fluoxetine. Journal of Clinical Neurophysiology, Klumpp, H., Keller, J., Miller, G.A., Casas, B.R., Best, J.L. &
28, 478–482. Deldin, P.J. (2010). Semantic processing of emotional words in
Hunter, A.M., Leuchter, A.F., Cook, I.A. & Abrams, M. (2010). depression and schizophrenia. International journal of Psycho-
Brain functional changes (QEEG cordance) and worsening sui- physiology, 75, 211–215.
cidal ideation and mood symptoms during antidepressant treat- Knott V.J. & Lapierre, Y.D. (1987). Computerized EEG correlates
ment. Acta Psychiatrica Scandinavica, 122, 461–469. of depression and antidepressant treatment. Progress in Neuro-
Iosifescu, D.V., Greenwald, S., Devlin, P., Mischoulon, D., Psychopharmacology & Biological Psychiatry, 11, 213–221.
Denninger, J.W., Alpert, J.E. & Fava, M. (2009). Frontal EEG Knott, V.J., Mahoney, C., Kennedy, S. & Evans, K. (2000). Pre-
predictors of treatment outcome in major depressive disorder. treatment EEG and its relationship to depression severity and
European Neuropsychopharmacology, 19, 772–777. paroxetine treatment outcome. Pharmacopsychiatry, 33, 201–205.
Işıntaş, M., Ak, M., Erdem, M., Oz, O. & Ozgen, F. (2012). Knott, V.J., Mahoney, C., Kennedy, S. & Evans, K. (2001). EEG
Majör Depresif Bozuklukta Olaya İlişkin Potansiyeller: Tedaviye power, frequency, asymmetry and coherence in male depres-
Yanıt ile P300 Arasındaki İlişki. [Event-related potentials in sion. Psychiatry Research, 106, 123–140.
major depressive disorder: The relationship between P300 and Knott, V.J, Telner, J.I., Lapierre, Y.D., Browne, M. & Horn, E.R.
treatment response]. Türk psikiyatri dergisi [Turkish Journal of (1996). Quantitative EEG in the prediction of antidepressant
Psychiatry], 23, 33–39. response to imipramine. Journal of Affective Disorders 39,
Itil, T. (1983). The significance of quantitative pharmaco-EEG in 175–184.
the discovery and classification of psychotropic drugs. In Korb, A., Cook, I., Hunter, A. & Leuchter, A. (2008). Brain
W. Herrmann, (Ed.), Electroencephalography in Drug Research electrical source differences between depressed subjects and
(pp. 131–158). Stuttgart: Gustav Fischer. healthy controls. Brain Topography, 21, 138–146.
 616 S. Olbrich & M. Arns
Kuo, C.-C. & Tsai, J.-F. (2010). Cordance or antidepressant Luthringer, R., Minot, R., Toussaint, M., Calvi-Gries, F.,
treatment response (ATR) index? Psychiatry Research, 180, 60. Schaltenbrand, N. & Macher, J.P. (1995). All-night EEG
Kwon, J.S., Youn, T. & Jung, H.Y. (1996). Right hemisphere spectral analysis as a tool for the prediction of clinical
abnormalities in major depression: Quantitative electroen- response to antidepressant treatment. Biological Psychiatry, 38,
cephalographic findings before and after treatment. Journal of 98–104.
Affective Disorders, 40, 169–173. Mayberg, H.S., Brannan, S.K., Mahurin, R.K., Jerabek, P.A.,
Landsness, E.C., Goldstein, M.R., Peterson, M.J., Tononi, G. & Brickman, J.S., Tekell, J.L., .. Fox, P.T. (1997). Cingulate func-
Benca, R.M. (2011). Antidepressant effects of selective tion in depression: A potential predictor of treatment response.
slow wave sleep deprivation in major depression: A high-density Neuroreport, 8, 1057–1061.
EEG investigation. Journal of Psychiatric Research, 45, Mientus, S., Gallinat, J., Wuebben, Y., Pascual-Marqui, R.D.,
1019–1026. Mulert, C., Frick, K., … Winterer, G. (2002). Cortical hypoac-
Lauer, C.J., Riemann, D., Wiegand, M. & Berger, M. (1991). tivation during resting EEG in schizophrenics but not in depres-
From early to late adulthood. Changes in EEG sleep of sives and schizotypal subjects as revealed by low resolution
depressed patients and healthy volunteers. Biological Psychiatry, electromagnetic tomography (LORETA). Psychiatry Research:
29, 979–993. Neuroimaging, 116, 95–111.
Lee, T.W., Yu, Y.W.Y., Chen, M.-C. & Chen, T.-J. (2011). Cortical Modell, S., Ising, M., Holsboer, F. & Lauer, C.J. (2005). The
Munich vulnerability study on affective disorders: Premorbid
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

mechanisms of the symptomatology in major depressive

disorder: A resting EEG study. Journal of Affective Disorders, 131, polysomnographic profile of affected high-risk probands. Bio-
243–250. logical Psychiatry, 58, 694–699.
Lee, T.-W., Yu, Y.W.Y., Chen, T.J. & Tsai, S.J. (2005). Loudness Morehouse, R.L., Kusumakar, V., Kutcher, S.P., LeBlanc, J. &
dependence of the auditory evoked potential and response to Armitage, R. (2002). Temporal coherence in ultradian sleep
antidepressants in Chinese patients with major depression. EEG rhythms in a never-depressed, high-risk cohort of female
Journal of Psychiatry and Neuroscience, 30, 202–205. adolescents. Biological Psychiatry, 51, 446–456.
Lemere, F. (1936). The significance of individual differences in the Mulert, C, Juckel, G., Augustin, H. & Hegerl, U. (2002). Com-
Berger rhythm. Brain, 59, 366–375. parison between the analysis of the loudness dependency of the
Leuchter, A.F., Cook, I.A., Gilmer, W.S., Marangell, L.B., auditory N1/P2 component with LORETA and dipole source
Burgoyne, K.S., Howland, R.H., … Greenwald, S. (2009). analysis in the prediction of treatment response to the selective
serotonin reuptake inhibitor citalopram in major depression.
Effectiveness of a quantitative electroencephalographic biomar-
Clinical Neurophysiology, 113, 1566–1572.
ker for predicting differential response or remission with
Mulert, C., Juckel, G., Brunnmeier, M., Karch, S., Leicht, G.,
For personal use only.

escitalopram and bupropion in major depressive disorder. Psy-

Mergl, R., … Pogarell, O. (2007). Prediction of treatment
chiatry Research, 169, 132–138.
response in major depression: Integration of concepts. Journal
Leuchter, A.F., Cook, I.A., Hunter, A.M., Cai, C. & Horvath, S.
of Affective Disorders, 98, 215–225.
(2012). Resting-state quantitative electroencephalography
Myslobodsky, M.S., Coppola, R., Bar-Ziv, J., Karson, C., Daniel,
reveals increased neurophysiologic connectivity in depression.
D., van Praag, H. & Weinberger, D.R. (1989). EEG asym-
PLoS ONE, 7(2), e32508.
metries may be affected by cranial and brain parenchymal
Leuchter, A.F, Cook, I.A., Uijtdehaage, S.H., Dunkin, J.,
asymmetries. Brain Topography, 1, 221–228.
Lufkin, R.B., Anderson-Hanley, C., … Babaie, A. (1997). Brain
Naismith, S.L., Mowszowski, L., Ward, P.B., Diamond, K., Para-
structure and function and the outcomes of treatment for
dise, M., Kaur, M., … Hermens, D.F. (2012). Reduced tempo-
depression. Journal of Clinical Psychiatry, 58(S16), 22–31.
ral mismatch negativity in late-life depression: An event-related
Leuchter, A.F., Cook, I.A., Witte, E.A., Morgan, M. &
potential index of cognitive deficit and functional disability?
Abrams, M. (2002). Changes in brain function of depressed Journal of Affective Disorders, 138, 71–78.
subjects during treatment with placebo. American Journal of Nissen, C., Feige, B., König, A., Voderholzer, U., Berger, M. &
Psychiatry, 159, 122–129. Riemann, D. (2001). Delta sleep ratio as a predictor of sleep
Leuchter, A.F., Uijtdehaage, S.H., Cook, I.A., O’Hara, R. & deprivation response in major depression. Journal of Psychiatric
Mandelkern, M. (1999). Relationship between brain electrical Research, 35, 155–163.
activity and cortical perfusion in normal subjects. Psychiatry Nyström, C., Matousek, M. & Hällström, T. (1986). Relationships
Research, 90, 125–140. between EEG and clinical characteristics in major depressive
Lieber, A.L. (1988). Diagnosis and subtyping of depressive disor- disorder. Acta Psychiatrica Scandinavica, 73, 390–394.
ders by quantitative electroencephalography: II. Interhemi- O’Connor, K.P., Shaw, J.C. & Ongley, C.O. (1979). The EEG and
spheric measures are abnormal in major depressives and differential diagnosis in psychogeriatrics. British Journal of Psy-
frequency analysis may discriminate certain subtypes. Hillside chiatry, 135, 156–162.
Journal of Clinical Psychiatry, 10, 84–97. Olbrich, S., Sander, C., Minkwitz, J., Chittka, T., Mergl, R.,
Lieber, A.L. & Prichep, L.S. (1988). Diagnosis and subtyping of Hegerl, U. & Himmerich, H. (2012). EEG vigilance regulation
depressive disorders by quantitative electroencephalography: I. patterns and their discriminative power to separate patients with
Discriminant analysis of selected variables in untreated depres- major depression from healthy controls. Neuropsychobiology, 65,
sives. Hillside Journal of Clinical Psychiatry, 10, 71–83. 188–194.
Linka, T., Sartory, G., Bender, S., Gastpar, M. & Müller, B.W. Park, C.A., Kwon, R.J., Kim, S., Jang, H., Chae, J.H., Kim, T. &
(2007). The intensity dependence of auditory ERP components Jeong, J. (2007). Decreased phase synchronization of the EEG in
in unmedicated patients with major depression and healthy patients with major depressive disorder. In R. Magjarevic & J.H.
controls. An analysis of group differences. Journal of Affective Nagel (Eds), World Congress on Medical Physics and Biomedical
Disorders, 103, 139–145. doi:10.1016/j.jad.2007.01.018 Engineering 2006 (pp. 1095–1098). Berlin: Springer. Retrieved
Lopes, M.C., Quera-Salva, M.-A. & Guilleminault, C. (2007). from http://link.springer.com/chapter/10.1007/978-3-540-36841-
Non-REM sleep instability in patients with major depressive dis- 0_262
order: Subjective improvement and improvement of non-REM Park, Y.M., Lee, S.H., Kim, S. & Bae, S.M. (2010). The loudness
sleep instability with treatment (agomelatine). Sleep Medicine, 9, dependence of the auditory evoked potential (LDAEP) in
33–41. schizophrenia, bipolar disorder, major depressive disorder,
 EEG biomarkers in major depressive disorder 617
anxiety disorder, and healthy controls. Progress in Neuro- marker of vulnerability to bipolar disorder and schizophrenia.
Psychopharmacology and Biological Psychiatry, 34, 313–316. Acta Psychiatrica Scandinavica, 117, 313–318.
Pizzagalli, D.A. (2011). Frontocingulate dysfunction in depres- Saletu, B., Anderer, P., Saletu-Zyhlarz, G.M., Gruber, D.,
sion: Toward biomarkers of treatment response. Neuropsychop- Metka, M. & Huber, J. (2005). Identifying target regions for
harmacology, 36, 183–206. vigilance improvement under hormone replacement therapy in
Pizzagalli, D.A., Oakes, T.R. & Davidson, R.J. (2003). Coupling postmenopausal syndrome patients by means of electroen-
of theta activity and glucose metabolism in the human rostral cephalographic tomography (LORETA). Psychopharmacology,
anterior cingulate cortex: An EEG/PET study of normal and 178, 389–399.
depressed subjects. Psychophysiology, 40, 939–949. Schaffer, C.E., Davidson, R.J. & Saron, C. (1983). Frontal
Pogarell, O., Padberg, F., Karch, S., Segmiller, F., Juckel, G., and parietal electroencephalogram asymmetry in depressed
Mulert, C., … Koch, W. (2011). Dopaminergic mechanisms of and nondepressed subjects. Biological Psychiatry, 18,
target detection – P300 event related potential and striatal 753–762.
dopamine. Psychiatry Research, 194, 212–218. Schmidt, H.D., Shelton, R.C. & Duman, R.S. (2011). Functional
Pozzi, D., Golimstock, A., Petracchi, M., García, H. & Starkstein, biomarkers of depression: Diagnosis, treatment, and patho-
S. (1995). Quantified electroencephalographic changes in physiology. Neuropsychopharmacology, 36, 2375–2394.
depressed patients with and without dementia. Biological Segrave, R.A., Cooper, N.R., Thomson, R.H., Croft, R.J., Shep-
Psychiatry, 38, 677–683.
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

pard, D.M. & Fitzgerald, P.B. (2011). Individualized alpha

Price, G.W., Lee, J.W., Garvey, C. & Gibson, N. (2008). Appraisal activity and frontal asymmetry in major depression. Clinical
of sessional EEG features as a correlate of clinical changes in EEG and Neuroscience, 42, 45–52.
an rTMS treatment of depression. Clinical EEG and Neuro- Smit, D.J., Posthuma, D., Boomsma, D.I. & De Geus, E.J. (2007).
science, 39, 131–138. The relation between frontal EEG asymmetry and the risk for
Prichep, L.S. & John, E.R. (1992). QEEG profiles of psychiatric anxiety and depression. Biological Psychology, 74, 26–33.
disorders. Brain Topography, 4, 249–257. Spronk, D., Arns, M., Barnett, K., Cooper, N. & Gordon, E.
Quera Salva, M.-A., Vanier, B., Laredo, J., Hartley, S., (2011). An investigation of EEG, genetic and cognitive markers
Chapotot, F., Moulin, C., … Guilleminault, C. (2007). Major of treatment response to antidepressant medication in patients
depressive disorder, sleep EEG and agomelatine: An open-label with major depressive disorder: A pilot study. Journal of Affective
study. International Journal of Neuropsychopharmacology, 10,
Disorders, 128, 41–48.
691–696.
Spronk, D.B., Veth, C.P., Arns, M., Schofield, P.R.,
Quinn, C.R., Harris, A. & Kemp, A.H. (2012). The impact of
Dobson-Stone, C., Ramaekers, J.G., … Verkes, R.J. (2013).
depression heterogeneity on inhibitory control. Australian and
For personal use only.

DBH-1021C⬎T and COMT Val108/158Met genotype are not

New Zealand Journal of Psychiatry, 46, 374–383.
associated with the P300 ERP in an auditory oddball task.
Rabinoff , M., Kitchen, C.M., Cook, I. & Leuchter, A. (2011).
Clinical Neurophysiology: official journal of the International
Evaluation of quantitative EEG by classification and regression
Federation of Clinical Neurophysiology, 124, 909–915.
trees to characterize responders to antidepressant and placebo
Suffin, S.C. & Emory, W.H. (1995). Neurometric subgroups in
treatment. Open Medical Informatics Journal, 5, 1–8.
attentional and affective disorders and their association with
Rao, U., Hammen, C.L. & Poland, R.E. (2009). Risk markers for
pharmacotherapeutic outcome. Journal of Clinical Electroen-
depression in adolescents: Sleep and HPA measures. Neuropsy-
cephalography, 26, 76–83.
chopharmacology, 34, 1936–1945.
Sun, Y., Li, Y., Zhu, Y., Chen, X. & Tong, S. (2008). Electroen-
Reid, S.A., Duke, L.M. & Allen, J.J. (1998). Resting frontal elec-
cephalographic differences between depressed and control sub-
troencephalographic asymmetry in depression: Inconsistencies
suggest the need to identify mediating factors. Psychophysiology, jects: An aspect of interdependence analysis. Brain Research
35, 389–404. Bulletin, 76, 559–564.
Reynolds, C.F., III, Kupfer, D.J., Taska, L.S., Hoch, C.C., Taylor, B.P., Bruder, G.E., Stewart, J.W., McGrath, P.J.,
Spiker, D.G., Sewitch, D.E., … Martin, D. (1985). EEG sleep Halperin, J., Ehrlichman, H. & Quitkin, F.M. (2006). Psycho-
in elderly depressed, demented, and healthy subjects. Biological motor slowing as a predictor of fluoxetine nonresponse in
Psychiatry, 20, 431–442. depressed outpatients. American Journal of Psychiatry, 163,
Roemer, R.A., Shagass, C., Dubin, W., Jaffe, R. & Siegal, L. 73–78.
(1992). Quantitative EEG in elderly depressives. Brain Topogra- Tenke, C.E., Kayser, J., Manna, C.G., Fekri, S., Kroppmann, C.J.,
phy, 4, 285–290. Schaller, J.D., . Bruder, G.E. (2011). Current source density
Rotenberg, V.S., Shami, E., Barak, Y., Indursky, P., Kayumov, L. measures of electroencephalographic alpha predict anti-
& Mark, M. (2002). REM sleep latency and wakefulness depressant treatment response. Biological Psychiatry, 70,
in the first sleep cycle as markers of major depression: A 388–394.
controlled study vs. schizophrenia and normal controls. Progress Tomarken, A.J., Davidson, R.J., Wheeler, R.E. & Kinney, L.
in Neuro-Psychopharmacology and Biological Psychiatry, 26, (1992). Psychometric properties of resting anterior EEG asym-
1211–1215. metry: Temporal stability and internal consistency. Psychophys-
Roth, W.T., Pfefferbaum, A., Kelly, A.F., Berger, P.A. & Kopell, iology, 29, 576–592.
B.S. (1981). Auditory event-related potentials in schizophrenia Tremblay, S., Beaulé, V., Proulx, S., de Beaumont, L.,
and depression. Psychiatry Research, 4, 199–212. Marjanska, M., Doyon, J., ..Théoret, H. (2013). Relationship
Ryu, V., An, S.K., Ha, R.Y., Kim, J.A., Ha, K. & Cho, H.-S. between transcranial magnetic stimulation measures of
(2012). Differential alteration of automatic semantic processing intracortical inhibition and spectroscopy measures of GABA
in treated patients affected by bipolar mania and schizophrenia: and glutamate⫹ glutamine. Journal of Neurophysiology, 109,
An N400 study. Progress in Neuro-Psychopharmacology and Bio- 1343–1349.
logical Psychiatry, 38, 194–200. Ulrich, G. & Fürstenberg, U. (1999). Quantitative assessment of
Sánchez-Morla, E.M., García-Jiménez, M.A., Barabash, A., dynamic electroencephalogram (EEG) organization as a tool
Martínez-Vizcaíno, V., Mena, J., Cabranes-Díaz, J.A., … for subtyping depressive syndromes. European Psychiatry, 14,
Santos, J.L. (2008). P50 sensory gating deficit is a common 217–229.
 618 S. Olbrich & M. Arns
Ulrich, G., Renfordt, E., Zeller, G. & Frick, K. (1984). Interrela- and its variability over time and their validity in differentiating
tion between changes in the EEG and psychopathology under healthy volunteers from depressed patients. Advances in Bio-
pharmacotherapy for endogenous depression. A contribution to logical Psychiatry, 13, 172–181.
the predictor question. Pharmacopsychiatry, 17, 178–183. Wang, Y., Fang, Y.-R., Chen, X.-S., Chen, J., Wu, Z.-G.,
Van Beijsterveldt, C.E. & van Baal, G.C. (2002). Twin and family Yuan, C.-M., … Cao, L. (2009). A follow-up study on features
studies of the human electroencephalogram: A review and a of sensory gating P50 in treatment-resistant depression patients.
meta-analysis. Biological Psychiatry, 61, 111–138. Chinese Medical Journal, 122, 2956–2960.
Vandoolaeghe, E., van Hunsel, F., Nuyten, D. & Maes, M. (1998). Williams, L.M., Rush, A.J., Koslow, S.H., Wisniewski, S.R.,
Auditory event related potentials in major depression: Pro- Cooper, N.J., Nemeroff , C.B., … Gordon, E. (2011). Inter-
longed P300 latency and increased P200 amplitude. Journal of national Study to Predict Optimized Treatment for Depres-
Affective Disorders, 48, 105–113. sion (iSPOT-D), a randomized clinical trial: rationale and
Volkert, J., Schulz, H., Härter, M., Wlodarczyk, O. & Andreas, S. protocol. Trials, 12, 4.
(2013). The prevalence of mental disorders in older people in Zoon, H.F.A., Veth, C.P.M., Arns, M., Drinkenburg, W.H.I.M.,
western countries – A meta-analysis. Ageing Research Reviews, Talloen, W., Peeters, P.J. & Kenemans, J.L. (2013). EEG alpha
12, 339–353. power as an intermediate measure between brain-derived neu-
Von Knorring, L., Perris, C., Goldstein, L., Kemali, D., rotrophic factor Val66Met and depression severity in patients
Monakhov, K., Vacca, L. (1983). Intercorrelations between dif- with major depressive disorder. Journal of Clinical Neurophysiology,
Int Rev Psychiatry Downloaded from informahealthcare.com by University of Connecticut on 01/05/14

ferent computer-based measures of the EEG alpha amplitude 30, 261–267.

For personal use only.