Clinical EEG and Neuroscience

Clinical EEG
and Neuroscience
http://eeg.sagepub.com/
The Quantitative Electroencephalogram and the Low-Resolution Electrical Tomographic Analysis in
Posttraumatic Stress Disorder
Doran Todder, Joseph Levine, Ahmad Abujumah, Michael Mater, Hagit Cohen and Zeev Kaplan
Clin EEG Neurosci 2012 43: 48
DOI: 10.1177/1550059411428716
The online version of this article can be found at:
http://eeg.sagepub.com/content/43/1/48
Published by:
http://www.sagepublications.com
On behalf of:
EEG and Clinical Neuroscience Society
Additional services and information for Clinical EEG and Neuroscience can be found at:
Email Alerts: http://eeg.sagepub.com/cgi/alerts
Subscriptions: http://eeg.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
>> Version of Record - Jan 1, 2012
What is This?
Downloaded from eeg.sagepub.com by guest on February 20, 2013
The Quantitative Electroencephalogram
and the Low-Resolution Electrical
Tomographic Analysis in Posttraumatic
Stress Disorder
Clinical EEG and Neuroscience
43(1) 48-53
ª EEG and Clinical Neuroscience
Society (ECNS) 2012
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1550059411428716
http://eeg.sagepub.com
Doran Todder1, Joseph Levine1, Ahmad Abujumah1, Michael Mater1,
Hagit Cohen1, and Zeev Kaplan1
Abstract
The electroencephalogram (EEG) is the recording of the brain electrical activity as measured on the scalp. Using mathematical
algorithms, the 3-dimensional (3D) distribution of the electrical potential inside the brain can be calculated. One of the methods
to calculate it is the low-resolution electrical tomographic analysis (LORETA). In this research, we seek to find the brain structures that differentiate patients with posttraumatic stress disorder (PTSD) from controls. Ten right-handed consenting adult male
patients were recruited from a PTSD clinic. All patients fulfilled Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition,
Text Revision [DSM-IV-TR]) criteria for chronic PTSD (duration >2 years.) and were on drug treatment regimens that had been
stable for at least 2 months (involving only serotonin reuptake inhibitors [SSRIs] and benzodiazepines).The control group consisted of 10 healthy hospital staff members. All study participants underwent 19 channel EEG measurements according to current
standards of practice. All artifact-free EEG strips were examined for spectral as well as LORETA analysis focusing on the theta (4-7
Hz) band which is suggested to reflect the activity of the limbic system. The theta band showed a statistically significant difference
(P < .05) between the 2 groups in the right temporal lobe and in both the right and left frontal lobes. Our findings support existing
research data obtained via other imaging technologies, which demonstrated structural alterations in the right temporal and frontal
areas in PTSD. These results indicate that combining quantitative EEG (QEEG) and the LORETA method, among other methods,
may improve the neuroanatomical resolution of EEG data analysis.
Keywords
electoencephalography, low-resolution electrical tomographic analysis, posttraumatic stress disorder
Received August 10, 2010; accepted June 6, 2011.
Introduction
PTSD affects about 9% of the general population and is clinically defined by the causal exposure to a significantly stressful experience arousing a certain pattern of response and
resulting in a specified number of symptoms of intrusive reexperiencing numbed emotions and avoidance of stimuli, and
of hyperarousal, present for at least 1 month after the event.
The disorder is often chronic and extremely incapacitating,
with long lasting psychological and physiological changes.1
However, whereas almost all of those exposed to the causal
events initially display an acute stress response, the majority
do not go on to develop the full-blown chronic clinical syndrome and apparently cope and readjust adequately. Improved
means for characterizing the distinctions between patients
with PTSD and other populations might reveal possible means
for early identification of higher-risk participants and enable
the development of more effective methods of primary and
secondary prevention.
PTSD has been the subject of numerous studies examining
the underlying functional neuroanatomy of the clinical symptoms, generally by means of functional magnetic resonance
imaging (fMRI). Most studies have focused on the response
to trauma-related versus other stimuli/challenges.2,3 Procedures such as the fMRI are costly and complex to perform and
have relatively poor time resolution. The EEG on the other
1
Ben Gurion University, Faculty of Health, Beer Sheva, Israel
Corresponding Author:
Doron Todder, Beer Sheva Mental Health Centre, Hazzadik Miroshalim, Beer
Sheva, Israel 84170
Email: dtoder@netvision.net.il
Downloaded from eeg.sagepub.com by guest on February 20, 2013
Todder et al.
49
hand offers a potentially valuable complementary source of
information for researchers and clinicians, since it assesses
real-time electrical activity in the brain and is overall a less
costly, time-consuming, and complex procedure. The raw EEG
data can be analyzed by direct visual inspection or by computerized quantitative EEG analysis (QEEG). The QEEG is
superior to the raw EEG analysis due to its larger reliability
and validity.
A small number of studies have explored QEEG patterns in
patients with PTSD and have elicited conflicting results. Begic
et al4 compared 18 unmedicated PTSD veterans to 20 control
participants. They found that the patients with PTSD have both
an increased theta power over central brain regions, as well as
increased beta activity over frontal, central, and occipital brain
regions. No significant differences between the groups were
noted for delta and alpha activity. In another study, the same
group compared veterans with PTSD to veterans without
PTSD. Patients with PTSD displayed predominantly
decreased alpha and increased beta power over frontal, central, and temporal areas. No difference was noted for the theta
rhythm between these groups.5 Rabe et al studied the hemispheric asymmetries among 4 groups. Three of the groups
were unmedicated motor vehicle survivors: Full-blown
PTSD; subsyndromal PTSD; no PTSD. The participants of the
fourth group were healthy controls. No difference was
noted between the groups for EEG alpha activity at rest.6
Metzger et al7 measured the alpha symmetry in a group of
female veterans with and without current PTSD. They found
some correlation between arousal symptoms and higher alpha
activity of the right compared to the left parietal lobe.7
Finally, Shankman et al8 compared the resting EEG of
patients with PTSD to highly selected control participants
with no history of reported trauma. No statistically significant
difference was noted on any of the spectral bands between the
groups.
Thus, the existing literature describing QEEG in patients with
PTSD shows no consistent trend in findings. With regard to the
theta band, although one report suggested that patients with
PTSD at rest may demonstrate increased theta activity over
central brain regions,4 another study failed to find such
differences.8
One of the possible mathematical transformations of EEG
data is the low-resolution electrical tomographic analysis
(LORETA, see9,10). LORETA uses measurements of the voltage potential over the scalp (raw EEG data) and then estimates
the current sources inside the brain that produce the measured
signal. This procedure helps to determine the relative contribution of different neuroanatomical structures to the measured
EEG. LORETA and other signal processing methods have the
benefit of superior time resolution of EEG measurements of
milliseconds, which is 3-fold better than that of fMRI, with
spatial resolution of approximately 7 mm, which is similar to
that of fMRI.11–13
Since it is theoretically possible that 2 EEG measurements
with similar visual raw EEG and/or QEEG patterns may differ
as to the inner brain electrical dipole distribution, the following
study will relate not only to the resting QEEG but also to the
LORETA analysis of EEG recordings of PTSD patients as
compared to controls.
Methods
Participants
Ten right-handed, consenting, adult male patients were
recruited from the PTSD clinic at the Beer Sheve Mental
Health Center (BS-MHC). All patients fulfilled the Diagnostic
and Statistical Manual of Mental Disorders (Fourth Edition,
Text Revision [DSM-IV-TR]) criteria1 for chronic PTSD (duration >2 years) were on drug treatment regimens which had been
stable for at least 2 months (involving only serotonin reuptake
inhibitor [SSRI] and benzodiazepines). None had a history of
neurological disorder, head trauma, alcohol or substance abuse,
or a previous psychotic episode. An equal number of healthy
male age-matched volunteer staff members acted as control
participants. The study was approved by the ethics committee
of the Ben-Gurion University of the Negev.
Procedure
The EEG was performed at rest, in a comfortable, quiet, airconditioned room, for 3 minutes with closed eyes, employing
the DeyMed Truescan 32 system with 19 electrodes (http://
www.deymed.com), arranged according to the 10 to 20 international conventions, with FPz as a reference. Bipolar eye movement electrodes were applied to the canthus and cheek bone on
the right eye in order to monitor the eye movement artifacts.
The amplifier bandwidths were nominally 1 to 40 Hz, and the
EEG was digitized at 128 Hz. During measurements, the impedance of all electrodes was kept below 5 kO.
Analysis of Data
Both split-half and test–retest reliability tests were conducted
on the edited, artifact-free, EEG segments. Only records with
>95% split half reliability and >90% test–retest reliability and
a total measurement of more than 1 minute were subjected to
the spectral and LORETA analyses. After editing, the shortest
EEG length was 1 minute, while the longest was approximately
2 minutes. The removal of artifacts and the calculation of the
statistical properties of the segments were performed using
NeuroGuide software (http://www.appliedneuroscience.com).
The artifacts removal were done both by the automatic algorithms in the NeuroGuide software and by visual inspection.
In order to avoid statistical multiple comparisons, the study
focused on one spectral band, namely the absolute theta band, a
band suggested to originate mainly from the limbic system,14
which has previously been reported to play a key role in PTSD
symptomatology.15
The spectral analysis was done using both the linked ears
reference montage and the local average (Laplacian) montage.
Since most QEEG norms were measured with the linked
ears montage, this was used as the standard. In addition, the
Downloaded from eeg.sagepub.com by guest on February 20, 2013
50
Clinical EEG and Neuroscience 43(1)
Table 1. Comparisons Between PTSD and Control Groups on Both Montagesa
Mean PTSD Group
Theta Distribution
Counter Map
Mean control Group
Theta Distribution
Counter Map
Statistically
Significant
Difference
Mean PTSD
Group Theta
Distribution (SD)
Mean Control
Group Theta
Distribution (SD)
Link ears montage (mV)
8.4 (2.2)
11.8 (3)
NS
Laplacian montage (mA)
221 (61)
434 (149)
NS
Abbreviations: PTSD, posttraumatic stress disorder; SD, standard deviation; NS, not significant.
a
The brain maps represent the mean theta distribution over the 19 electrodes for both groups.
Laplacian montage was employed due to its ability to highlight
local fields14 and the fact that it has been suggested to minimize
global effects on the brain that drugs usually do.
Statistics
As stated before, in order to avoid multiple comparisons only
absolute theta band values were compared. The unpaired t test
was applied for comparing the data of both the QEEG spectral
analysis and the LORETA analysis. Since the EEG strip was
divided to 2 second epochs, there were over 30 degrees of freedom and therefore T > 2.1 stands for alpha level set at .05.
There was no statistically significant age difference between
the study groups (P ¼ .19; patients with PTSD—mean age:
44.4, SD: 9 years vs control participants —mean age: 39.1,
SD: 9 years).
Results
On the QEEG, no statistically significant difference was found
between PTSD and control groups for the theta band (4-8 Hz)
on both montages (see Table 1). The LORETA analysis, however, revealed distinct patterns for patients with PTSD as compared to controls. Patients with PTSD displayed statistically
significantly lower activity on the ‘‘low’’ theta band (4-5 Hz),
mainly over the right temporal lobe (including Brodmann 13,
20, 21, 22, and 42 areas; Figure 1). On the ‘‘higher’’ theta band
(6-7 Hz), patients with PTSD demonstrated lower activity over
both the right and left frontal lobes (including Brodmann 9,
10, 44, 45, and 46 areas; Figure 2).
Discussion
This study employed 2 methods of brain EEG analysis,
namely QEEG and LORETA, in patients with PTSD compared to controls. No statistically significant difference was
noted between these groups for the QEEG theta band. However, a statistically significant difference between PTSD and
controls was found in the LORETA analysis, particularly on
the right temporal lobe and both the right and left frontal
lobes.
The significant findings on the LORETA analysis are
important. There are relatively few studies reporting brain neuroimaging data at rest, while there is a large number of studies
applying provocation paradigms prior to, or at the time of the
imaging procedure.16 In an fMRI study, Lucey et al17 found
that patients with PTSD differ from control participants on the
bilateral superior frontal cortices and on the right caudate at
rest. Our study also identified different activation of the superior frontal region (Brodmann 9 and 10) between patients with
PTSD and controls. The caudate is generally not considered
to be an EEG-producing structure and therefore is not a part
of the Talairach coordinates serving as the base for the LORETA
neuroanatomical identified structures. (http://www.uzh.ch/keyinst/NewLORETA/TechnicalDetails/TechnicalDetails.htm).
Therefore, the functioning of this structure cannot be
explored by LORETA.
Even if one considers the brain regions relevant to PTSD of
these patients while not at rest (ie, subjected to a variety of
cognitive tasks), the right temporal as well as the right and left
frontal brain areas are reported to differ between PTSD
Downloaded from eeg.sagepub.com by guest on February 20, 2013
Todder et al.
51
Areas of statistically significant difference between PTSD and control
groups
LORETA
analysis
5 Hz
Figure 1. Group difference between PTSD and controls, in the LORETA analysis. The dark gray areas represent locations with statistically
significant differences (T 2.1) between the study groups. In this figure, the 5 Hz results are shown representing the difference in the lower
theta band (4-5 Hz). The LORETA analysis showed a statistically significant lower activity on the ‘‘low’’ theta band (4-6 Hz) in the PTSD group
compared to controls, mainly over the right temporal lobe (including Brodmann 13, 20, 21, 22, and 42 areas). PTSD indicates posttraumatic
stress disorder; LORETA, low-resolution electrical tomographic analysis.
patients and controls.18–21 The affected frontal structures
were linked previously to the experience and regulation of
emotion,22 while the right temporal lobe dysfunction can elicit the re-enactment and re-living of past experiences that may
cause the flashback symptoms.23
One of the current neural models for PTSD is associated
with deficits in information processing as well as with hypervigilance to salient and threat-related stimuli.19 This model combines 2 neural processes: hyperarousal on one hand, and the
breakdown of an inhibitory function required for attention control and working memory on the other. Evidence for this model
came recently from Falconer et al.24 These authors reported
that compared to controls, patients with PTSD activated mostly
the left ventrolateral prefrontal brain cortex, while a reduced
activation of the right frontotemporoparietal cortical inhibitory
network was also noted. Therefore, our results seem to be in
agreement with the above model, assuming a compromised
cortical inhibitory control brain network in PTSD.
The QEEG findings are consistent with those of Shankman
et al8 who reported no difference between patients with PTSD
and control participants at rest but are in contrast with Begic
et al4 who reported a significant difference for the theta band
between patients with PTSD and controls.
There are important differences between the above studies
that may account for the conflicting results. Posttraumatic
stress disorder is a heterogeneous disorder, and the nature of the
traumatic event may be of varying severity and/or expression
of symptoms.25 In the study by Shankman et al,8 different kinds
of events were included, while in the Begic et al study4 all
patients with PTSD were veterans, and it is reasonable to
Downloaded from eeg.sagepub.com by guest on February 20, 2013
52
Clinical EEG and Neuroscience 43(1)
LORETA
analysis
7Hz
Figure 2. Group difference between PTSD and controls, in the LORETA analysis. The dark gray areas represent locations with statistically
significant differences (T 2.1) between the study groups. In this figure, the 7 Hz results are shown representing the difference in the higher
theta band (6-7 Hz). The LORETA analysis showed a statistically significant lower activity in the ‘‘higher’’ theta band (6-7 Hz), for patients with
assume that these participants were exposed to a more limited
spectrum of events.
Psychotropic drug treatment may also induce changes in the
QEEG. In the study by Begic et al,4 all patients were drug free for
at least 2 weeks, while in the study by Shankman et al8 there was
no specific mention of drug therapy, and one may assume that the
patients were medicated. A third possible difference between the
studies is that the patients evaluated by Begic et al4 were not classified according to the clinical clusters of PTSD (intrusive vs
avoidance vs arousal clusters), while in Shankman et al8 there was
an attempt to correlate the QEEG findings with these clusters as
well as with certain personality traits. Another difference is the
diagnosis procedure. Begic et al used the DSM-IV4 while Shankman et al relied on clinical interview and the Clinician-Administered PTSD Scale (CAPS) questionnaire.8
While the patients in our study were veterans diagnosed
according to the DSM-IV-TR criteria, similar to those in the
study by Begic et al, they were receiving medications, in
contrast to those participants studied by Begic et al. One may
postulate that the lack of drug treatment use by Begic et al4
contributed to their positive findings. However, SSRIs are usually considered to influence the alpha band, and benzodiazepines
are usually considered to influence the beta band, rather than the
theta band,14 which is our focus. Therefore, the use of psychotropic medications may thus not explain the difference between our
findings and those of Begic et al.4 More research is needed in
this regard.
Our study has several limitations: (a) this study consisted of
a relatively small number of patients; (b) patients were treated
with SSRIs and benzodiazepines and thus were not drug free.
On the other hand, the effect of medications is usually considered to cause global changes, while our results show focal
affects on the brain, thus making a drug effect less plausible.
(c) a state of ‘‘rest’’ is not the same as a ‘‘relaxed’’ state. Some
Downloaded from eeg.sagepub.com by guest on February 20, 2013
Todder et al.
53
of our patients may have experienced anxiety due to a new and
unfamiliar examination, although they denied that this provoked anxiety. In the future, objective physiological measurement of arousal may well be applied during the rest condition.
Finally, this research demonstrates the importance of the
exploration of the electrical dipole distribution inside the brain.
In this regard, McFarlane et al,26 in a review article describing
the neural network model in PTSD, emphasized the importance
of short-time resolution brain mapping, since neural function is
in the millisecond range, while hemodynamic base brain mapping like fMRI operates on several seconds resolution. The
development of EEG analyses such as LORETA could thus
provide us with an important brain mapping technique using
adequate time and space resolution, which are needed for the
study of the neural network function in patients with PTSD.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to
the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship,
and/or publication of this article.
References
1. APA. Diagnostic and Sstatistical Manual of Mental Disorders.
Washington DC: APA; 2000.
2. Francati V, Vermetten E, Bremner JD. Functional neuroimaging
studies in posttraumatic stress disorder: review of current methods
and findings. Depress Anxiety. 2007;24(3):202–218.
3. Liberzon I, Sripada CS. The functional neuroanatomy of PTSD: a
critical review. Prog Brain Res. 2008;167:151–169.
4. Begic D, Hotujac L, Jokic-Begic N. Electroencephalographic comparison of veterans with combat-related post-traumatic stress disorder and healthy subjects. Int J Psychophysiol. 2001;40(2):167–172.
5. Jokic-Begic N, Begic D. Quantitative electroencephalogram
(qEEG) in combat veterans with post-traumatic stress disorder
(PTSD). Nord J Psychiatry. 2003;57(5):351–355.
6. Rabe S, Beauducel A, Zollner T, Maercker A, Karl A. Regional
brain electrical activity in posttraumatic stress disorder after
motor vehicle accident. J Abnorm Psychol. 2006;115(4):687–698.
7. Metzger LJ, Paige SR, Carson MA, et al. PTSD arousal and
depression symptoms associated with increased right-sided parietal EEG asymmetry. J Abnorm Psychol. 2004;113(2): 324–329.
8. Shankman SA, Silverstein SM, Williams LM, et al. Resting electroencephalogram asymmetry and posttraumatic stress disorder.
J Trauma Stress. 2008;21(2):190–198.
9. Pascual-Marqui RD. The spherical spline Laplacian does not
produce artifactually high coherences: comments on two articles
by Biggins et al. Electroencephalogr Clin Neurophysiol. 1993;
87(1):62–66.
10. Pascual-Marqui RD, Michel CM, Lehmann D. Low resolution
electromagnetic tomography: a new method for localizing electrical activity in the brain. Int J Psychophysiol. 1994;18(1):49–65.
11. Grech R, Cassar T, Muscat J, et al. Review on solving the inverse
problem in EEG source analysis. J Neuroeng Rehabil. 2008;5:25.
12. Baillet S, Garnero L. A Bayesian approach to introducing anatomofunctional priors in the EEG/MEG inverse problem. IEEE Trans
Biomed Eng. 1997;44(5):374–385.
13. Stern Y, Neufeld MY, Kipervasser S, et al. Source localization of
temporal lobe epilepsy using PCA-LORETA analysis on ictal
EEG recordings. J Clin Neurophysiol. 2009;26(2):109–116.
14. Niedermeyer E, Lopes da Silva F, eds. Electroencphalography.
5th ed. Baltimore MD: Lippincott Williams & Wilkins; 2005.
15. Bremner JD, Elzinga B, Schmahl C, Vermetten E. Structural and
functional plasticity of the human brain in posttraumatic stress
disorder. Prog Brain Res. 2008;167:171–186.
16. Lanius RA, Williamson PC, Bluhm RL, et al. Functional connectivity of dissociative responses in posttraumatic stress disorder: a
functional magnetic resonance imaging investigation. Biol Psychiatry. 2005;57(8):873–884.
17. Lucey JV, Costa DC, Adshead G, et al. Brain blood flow in
anxiety disorders. OCD, panic disorder with agoraphobia, and
post-traumatic stress disorder on 99mTcHMPAO single photon
emission tomography (SPET). Br J Psychiatry. 1997;171:
346–350.
18. Clark CR, Galletly CA, Ash DJ, et al. Evidence-based medicine
evaluation of electrophysiological studies of the anxiety disorders. Clin EEG Neurosci. 2009;40(2):84–112.
19. Falconer E, Bryant R, Felmingham KL, et al. The neural networks
of inhibitory control in posttraumatic stress disorder. J Psychiatry
Neurosci. 2008;33(5):413–422.
20. Lindauer RJ, Booij J, Habraken JB, et al. Effects of psychotherapy
on regional cerebral blood flow during trauma imagery in patients
with post-traumatic stress disorder: a randomized clinical trial.
Psychol Med. 2008;38(4):543–554.
21. Nutt DJ, Malizia AL. Structural and functional brain changes
in posttraumatic stress disorder. J Clin Psychiatry. 2004;
65(suppl 1):11–17.
22. Etkin A, Wager TD. Functional neuroimaging of anxiety: a
meta-analysis of emotional processing in PTSD, social anxiety
disorder, and specific phobia. Am J Psychiatry. 2007;164(10):
1476–1488.
23. Engdahl B, Leuthold AC, Tan HR, et al. Post-traumatic stress disorder: a right temporal lobe syndrome? J Neural Eng. 2010;7(6):
066005, 1–8.
24. Falconer EM, Felmingham KL, Allen A, et al. Developing an
integrated brain, behavior and biological response profile in posttraumatic stress disorder (PTSD). J Integr Neurosci. 2008;7(3):
439–456.
25. Lancaster SL, Melka SE, Rodriguez BF. An examination of
the differential effects of the experience of DSM-IV defined traumatic events and life stressors. J Anxiety Disord. 2009;23(5):
711–717.
26. McFarlane AC, Yehuda R, Clark CR. Biologic models of
traumatic memories and post-traumatic stress disorder: the role
of neural networks. Psychiatr Clin North Am. 2002;25(2):
253–270.
Downloaded from eeg.sagepub.com by guest on February 20, 2013