Dysfunctions of cortical excitability in drug-naïve posttraumatic stress disorder patients.
ABSTRACT The investigation of a wide set of transcranial magnetic stimulation (TMS)-related variables in both hemispheres might help to identify a pattern of cortical excitability changes in posttraumatic stress disorder (PTSD) patients, reflecting gamma-amino-butiric acid (GABA)/glutamate balance and dysfunction, and to determine whether some of these variables are related to clinical features.
In 20 drug-naive PTSD patients without comorbidity and 16 matched healthy control subjects we tested bilaterally with standard TMS procedures: resting motor threshold (RMT) to single-pulse TMS (reflecting ion channel function), paired-pulse short-latency intracortical inhibition (SICI; mainly reflecting GABA(A) function) and intracortical facilitation (ICF; mainly reflecting glutamatergic function), single-pulse cortical silent period (CSP; mainly reflecting GABA(B)-ergic function), and paired-pulse short-latency afferent inhibition (SAI; reflecting cholinergic mechanisms and their presynaptic GABA(A)-mediated modulation).
The PTSD patients showed widespread impairment of GABA(A)-ergic SICI, which was reversed toward facilitation in both hemispheres in one-half of the patients, marked increase of glutamatergic ICF in the right hemisphere, and right-sided impairment of SAI. Illness duration and avoidance symptoms but not anxiety correlated with right-lateralized dysfunctions of cortical excitability.
Although the neurobiological complexity of each TMS variable makes current results theoretical, the pattern of cortical excitability accompanying PTSD symptoms suggests a bilateral decrease of the GABA(A)-ergic function. This prevails in the right hemisphere, in association with a relative prevalence of the glutamatergic tone, a new finding that current neuroimaging investigations cannot provide due to the lack of reliable glutamate tracers. Results might help to disclose new pathophysiological aspects of PTSD symptoms, providing a rationale for future neuromodulatory strategies of treatment.
- [show abstract] [hide abstract]
ABSTRACT: Short-interval intracortical inhibition (SICI) is a widely used paired-pulse transcranial magnetic stimulation (TMS) measure to assess inhibition in human motor cortex. However, facilitatory processes may contaminate SICI under certain conditions. Here, we specifically address the contribution of short-interval intracortical facilitation (SICF). A SICF interstimulus interval (ISI) curve was obtained in nine healthy subjects according to an established paired-pulse TMS protocol [Ziemann U, Tergau F, Wassermann EM, Wischer S, Hildebrandt J, Paulus W. Demonstration of facilitatory I-wave interaction in the human motor cortex by paired transcranial magnetic stimulation. J Physiol (Lond) 1998a;511:181-190]. The individual ISI leading to SICF peak1, trough1, peak2, trough2 and peak3 was selected for the subsequent measurement of SICI intensity curves (SICI(peak1), SICI(trough1), SICI(peak2), SICI(trough2), SICI(peak3)) using intensity variation of the first stimulus (S1) from 50% to 120% of active motor threshold (AMT) in the first dorsal interosseous muscle. SICI(peak1) (mean ISI, 1.54ms) and SICI(trough1) (mean ISI, 1.97ms) showed a sigmoid SICI increase with S1 intensity. SICI(trough1) reached the strongest SICI and was therefore chosen for comparison with the other SICI curves. SICI(peak2) (mean ISI, 2.61ms) was U-shaped with a similar increase at low S1 intensities, but a decrease when S1 intensity exceeded 90% AMT. Correlation analyses suggested that this decrease was caused by SICF. SICI(trough2) (mean ISI, 3.50ms) and SICI(peak3) (mean ISI, 4.26ms) showed considerably less inhibition than SICI(trough1) over the whole range of S1 intensities. Findings show that commonly accepted protocols of testing SICI (ISI of 2-3ms, S1 intensity approximately 95% AMT) bear the risk of measuring net inhibition contaminated by SICF. SICF may contribute to apparently reduced SICI in patients with neurological or psychiatric disorders.Clinical Neurophysiology 09/2008; 119(10):2291-7. · 3.14 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Posttraumatic stress disorder (PTSD) is associated with long-term changes in neurobiology. Brain areas involved in the stress response include the medial prefrontal cortex, hippocampus, and amygdala. Neurohormonal systems that act on the brain areas to modulate PTSD symptoms and memory include glucocorticoids and norepinephrine. Dysfunction of these brain areas is responsible for the symptoms of PTSD. Brain imaging studies show that PTSD patients have increased amygdala reactivity during fear acquisition. Other studies show smaller hippocampal volume. A failure of medial prefrontal/anterior cingulate activation with re-experiencing of the trauma is hypothesized to represent a neural correlate of the failure of extinction seen in PTSD. The brain has the capacity for plasticity in the aftermath of traumatic stress. Antidepressant treatments and changes in environment can reverse the effects of stress on hippocampal neurogenesis, and humans with PTSD showed increased hippocampal volume with both paroxetine and phenytoin.Progress in brain research 02/2008; 167:171-86. · 4.19 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: The effect of the glutamate antagonist riluzole on excitatory and inhibitory phenomena in the human motor system was studied by transcranial magnetic stimulation (TMS) and peripheral electrical nerve stimulation. The motor threshold, the intracortical inhibition and intracortical facilitation as assessed by paired TMS, the cortical and peripheral silent periods, F wave amplitudes and F wave latencies were measured. Riluzole suppressed the intracortical facilitation whereas other parameters remained unchanged, indicating that the neurotransmitter glutamate is mainly involved in facilitatory mechanisms in the motor system.Journal of Neural Transmission 01/1997; 104(11-12):1207-14. · 3.05 Impact Factor
Dysfunctions of Cortical Excitability in Drug-Naïve
Posttraumatic Stress Disorder Patients
Simone Rossi, Alberto De Capua, Maricla Tavanti, Sara Calossi, Nicola R. Polizzotto, Antonio Mantovani,
Vincenzo Falzarano, Letizia Bossini, Stefano Passero, Sabina Bartalini, and Monica Ulivelli
to identify a pattern of cortical excitability changes in posttraumatic stress disorder (PTSD) patients, reflecting ?-amino-butiric acid
(GABA)/glutamate balance and dysfunction, and to determine whether some of these variables are related to clinical features.
cal inhibition (SICI; mainly reflecting GABAAfunction) and intracortical facilitation (ICF; mainly reflecting glutamatergic function), single-
pulse cortical silent period (CSP; mainly reflecting GABAB-ergic function), and paired-pulse short-latency afferent inhibition (SAI; reflecting
cholinergic mechanisms and their presynaptic GABAA-mediated modulation).
Results: The PTSD patients showed widespread impairment of GABAA-ergic SICI, which was reversed toward facilitation in both hemi-
duration and avoidance symptoms but not anxiety correlated with right-lateralized dysfunctions of cortical excitability.
Conclusions: Although the neurobiological complexity of each TMS variable makes current results theoretical, the pattern of cortical
rationale for future neuromodulatory strategies of treatment.
Key Words: GABA, glutamate, neuroimaging, neurophysiology,
in the pathophysiology of posttraumatic stress disorder (PTSD), a
disabling clinical syndrome including emotional numbing and
withdrawal, psychological and physiological hyperarousal, and
sleep and somatic disturbances (1). Animal data of PTSD-like
models have consistently shown decreased function of the
GABAAreceptors complex at cortical and hippocampal levels
(2,3), whereas socially isolated mice, exhibiting contextual fear
responses and impaired fear extinction, had decreased corti-
colimbic expression of allopreganolone, a positive allosteric
modulator of GABAAreceptors (4).
In a positron emission tomography study, a global reduction
of [11C]flumazenil binding was observed in veterans who devel-
oped PTSD versus veterans exposed to trauma but without
PTSD, thereby providing circumstantial evidence for the role of
the benzodiazepine-GABAAreceptor in the pathophysiology
of this disorder (5) and extending previous semiquantitative
studies performed with [123I]iomazenil single-photon emission
computed tomography that provided nonhomogeneous re-
sults (6,7). A recent investigation by transcranial magnetic
stimulation (TMS) suggests that GABA-mediated regulation of
nimal models and clinical studies converge in identifying a
possible dysfunction of the ?-amino-butiric acid (GABA)
cortical excitability might be impaired even when PTSD
symptoms are mild (8).
A limitation of these neuroimaging and neurophysiological
studies in PTSD is that only GABA dysfunctions have been
assessed, even if the regulation of cortical excitability relies on a
complex interplay of inhibitory and excitatory—mainly glutama-
tergic—presynaptic inputs acting ultimately on the fine regula-
tion of the excitability of corticospinal neurons. Moreover,
emerging views point also to a role of the glutamatergic system
as a neurobiological substrate underpinning the generation and
maintenance of PTSD symptoms, especially concerning stress-
and fear-related neuro-plastic synaptic changes, hyperarousal,
and re-experiencing of traumatic memories (9,10).
A large portion of these anatomic-physiologically distinct
circuitries acting on corticospinal neurons can be selectively
tested in vivo by single-pulse and paired-pulse TMS of the motor
Among single-pulse variables, the minimal intensity required
to produce a motor-evoked potential (MEP) with 50% probability
in a relaxed muscle (11) or resting motor threshold (RMT)
provides information about the central core of neurons of the
muscle representation within the motor cortex and reflects
membrane excitability, because it increases after administration
of drugs blocking voltage-gated sodium channels (12–14). The
suppression of the ongoing electromyographic (EMG) activity
due to the TMS pulse or cortical silent period (CSP) (15) reflects
the recruitment of cortical inhibitory (mainly GABAB-mediated)
interneurons in its later part (16–18), after an early phase of
spinal origin (15).
Paired-pulse measures of cortical excitability include the short
latency intracortical inhibition (SICI), the intracortical facilitation
(ICF), and the short latency afferent inhibition (SAI). Although
each TMS-related variable is mediated by complex neurobiolog-
ical mechanisms, there is general agreement that SICI—which is
From the Department of Neuroscience (SR, SP, SB, MU), Neurology Section;
and the Department of Neuroscience (ADC, MT, SC, NRP, AM, VF, LB),
Psychiatry Section, University of Siena School of Medicine, Siena, Italy.
Siena, Italy; E-mail: Rossisimo@unisi.it.
BIOL PSYCHIATRY 2009;66:54–61
© 2009 Society of Biological Psychiatry
enhanced by GABAergic drugs (13,19)—originates from low-
threshold GABAA-mediated inhibitory inputs (14,20,21) and that
ICF, which is reduced by N-methyl d-aspartate (NMDA) receptor
antagonists (22), is largely mediated by excitatory inputs from
high-threshold glutamatergic pathways to the motor cortex
(14,21,23). The SAI, which is decreased by muscarinic receptor
agonists in healthy subjects (24) and is enhanced by acetylcho-
linesterase inhibitors in patients with Alzheimer’s syndrome
(25), reflects mainly cholinergic cortical mechanisms, but
GABAA-mediated inhibition might also play a role in modulating
SAI at a presynaptic level (26).
With the exception of SICI (8), all these measures of cortical
excitability have not yet been explored in PTSD patients, despite
their potential usefulness in disclosing dysfunctions of neuro-
transmitters, such as glutamate, for which reliable positron
emission tomography tracers are under investigation but not yet
available for research and clinical applications (27).
Hence, the primary aim of the study was to determine
whether a pattern of cortical excitability dysfunction, on the basis
of a wide set of TMS-related variables derived from the two
hemispheres, can be identified in drug-naive PTSD patients. The
secondary goal was to determine whether some of these vari-
ables were related to clinical features.
Methods and Materials
From September 2007 to March 2008, 36 patients (female ?
20, male ? 16) with diagnosis of PTSD were initially screened at
the Centre for the Diagnosis and Treatment of PTSD, Department
of Neuroscience, University of Siena Medical Center. Psychiatric
diagnoses, on the basis of the Structured Clinical Interview
(SCID) for DSM-IV (1), were determined by a consensus of at
least two of the psychiatrists. Criteria of inclusion were age range
20–65 years, diagnosis of PTSD, and right-handedness. Criteria of
exclusion were a history of current and/or lifetime comorbid
psychiatric diagnoses on the basis of Axis I and Axis II interviews
with SCID-I/P (28) and SCID-II/P (29); previous use of any psych-
otropic medication; history of major head trauma; or presence of
neurological, endocrine, and degenerative disorders.
Ten patients were excluded because of the presence of a
psychiatric diagnosis other than PTSD (7 with major depressive
disorder, 3 with chronic alcoholism). Six patients with exclusive
diagnosis of PTSD were excluded because they were taking
psychotropic medications (sertraline or paroxetine). Therefore,
from the initial sample of 36, 20 patients were included in the
current study (female ? 10, male ? 10, mean age 50.6 ? 14.6
For comparison, 16 healthy right-handed subjects (female ?
7, male ? 9, mean age 40.8 ? 13.9 years) were recruited from the
general population of Siena and had neither lifetime nor current
histories of psychiatric disorders, including alcohol and sub-
stance use disorders as determined by the SCID Non-Patient
Version (30). Control subjects with a history of meningitis;
traumatic brain injury; presence of neurological, endocrine, or
degenerative disorders; use of drugs, or previous use of any
psychotropic medication were excluded. In neither group had
the women used oral contraception before this study.
Finally, both patients and control subjects denied having used
more than two cups/day of coffee or having used marijuana or
other substances in the last week preceding the neurophysiolog-
Written informed consent was obtained from all subjects after
a complete description of the study, which was approved by the
local ethics committee and was prepared in accordance with the
ethical standards laid down in the Declaration of Helsinki.
We obtained information about the traumatic events and ages
at which these events occurred with diagnostic interviews per-
formed by two different psychiatrists. The SCID-P for DSM-IV
was used to establish PTSD diagnosis.
Patients were assessed through the Clinician Administered
Posttraumatic Stress Disorder Scale (CAPS) (31). The CAPS gives
an indication of an individual’s symptoms severity, on three
scales (re-experiencing, avoidance and numbing, hyperarousal).
Furthermore, the evaluation for the presence of overlapping
symptoms between PTSD and major depressive disorder and
state of anxiety was also carried out with the Hamilton Depres-
sion Rating Scale (HAM-D) (32) and the Hamilton Anxiety Rating
Scale (HAM-A) (33), respectively.
Measures of Cortical Excitability by TMS: Acquisition and
Cortical excitability measures were collected in the late morn-
ing, at the last meal, or at coffee being taken in the early morning.
Smokers were asked to refrain from smoking in the 2 hours
preceding the test.
The TMS was applied via an eight-shaped focal coil connected
with two monophasic Magstim 200 stimulators controlled by a
Bi-Stim module (Magstim Company, Camarthenshire, United King-
dom). The coil, angled approximately 45° from the midline with its
handle pointing backward, was positioned on the scalp region (of
the left or right hemisphere) triggering MEPs from the contralat-
eral right and left first dorsal interosseous (FDI) hand muscles
with the minimal threshold (hot spot), as defined according to
international standards (11). Hot spots were marked on the scalp
to allow the same coil positioning during the whole session.
Silver–silver chloride (Ag–AgCl) adhesive electrodes were
applied over the muscles in a belly-tendon bipolar montage, with
the active electrode placed on the motor point of the FDI muscle.
The MEPs were recorded by a four-channel electromyograph
(Phasis, EBNeuro, Florence, Italy), with a bandpass filter of 20
Hz–5 kHz (sampling 20 kHz, gain range .01–1 mV, depending of
the variable tested). An acoustic feedback monitoring the EMG
background activity was given to subjects.
The following neurophysiological variables were bilaterally
collected, balancing their order and that of the studied hemi-
sphere, after having obtained individual RMTs.
RMT. With the arm fully relaxed, as indexed by the absence
of EMG activity, while the subjects fixed a point in front of them,
the minimal intensity (expressed as percentage of the maximal
stimulator output) required to obtain an MEP of approximately
50 ?V with 50% probability in at least 10 trials was searched.
SICI and ICF.
A conditioning subthreshold (80% of RMT)
pulse precedes the test pulse by 3 msec (SICI) or by 12 msec
(ICF), adjusted to obtain an MEP of approximately 600–800 ?V
(when unconditioned) in the resting FDI muscle (corresponding
to approximately 120% RMT) (20). Five trials were collected for
each interstimulus interval (ISI), with the order randomized. The
ISIs of 3 msec and 12 msec were considered instead of the full
range of possible ISIs, because these ISIs fall into the timing of
maximal expression of ICI (i.e., 2–4 msec) and ICF (i.e., 10–15
msec), respectively (14,20,34,35). The peak-to-peak amplitudes
S. Rossi et al.
BIOL PSYCHIATRY 2009;66:54–61 55
of the conditioned MEPs were averaged and expressed as a
percentage of the average of the test MEPs amplitude.
CSP. Subjects were asked to produce a voluntary maximal
contraction of the left or right FDI muscles. Ten single suprath-
reshold (130% of the RMT) TMS pulses spaced by at least 7 sec
were delivered on the contralateral motor cortex. These induced
a suppression of the underlying EMG activity. The length of the
CSP was determined on the rectified trace from the peak of the
MEP to the reappearance of the EMG activity (15). Mean values
of the 10 trials were considered for further analysis.
The SAI was performed in 14 PTSD patients and 15
control subjects. In accordance with the original paradigm (36),
the conditioning stimulus was an electric pulse delivered on the
ulnar nerve at wrist, at an intensity sufficient to produce a
nonpainful muscle twitch. The test stimulus was a suprathreshold
single TMS pulse delivered on the contralateral motor cortex
evoking, when unconditioned, a stable MEP in the FDI muscle of
approximately 600–800 ?V. The ISI between the two stimuli was
22 msec (i.e., 1 msec after the arrival of the ulnar nerve afferent
volley to the contralateral sensorimotor cortex). Only the “early”
conditioning was considered, because the “late” afferent volley
(i.e., approximately 30 msec after the peripheral stimulation)
enhances conditioned MEPs (37), and the underlying mecha-
nisms are not yet fully explored. The peak-to-peak amplitudes of
five conditioned MEPs were averaged and expressed as a
percentage of the average of five test MEPs amplitude.
The statistical analyses were performed with SPSS version
13.0 for Windows (SPSS, Chicago, Illinois). Comparisons be-
tween demographic variables of PTSD subjects and healthy
control subjects were analyzed with two-tailed unpaired t test or
two-tailed Mann–Whitney U test if the variables were nonnor-
mally distributed (as assessed with the Shapiro-Wilk test). Be-
tween-groups comparison of categorical variables was assessed
A repeated-measurements multivariate analysis of covariance
(ANCOVA) was conducted to compare TMS variables between
the two groups, with hemisphere (left vs. right) as the repeated
factor, group (PTSD and control subjects) and gender (female
and male) as the between-subjects factors, age as covariate, and
the TMS variables (RMT, SICI, ICF, CSP, SAI) as the within-subject
factors, followed by post hoc univariate ANCOVAs. Bonferroni
correction was applied whenever appropriate (i.e., for six com-
parisons). Hence, all effects are reported at p ? .0083.
Finally, clinical scores (illness duration, CAPS subscores of
re-experiencing, avoidance and hyper-arousal, scores of HAM-A
and HAM-D) and neurophysiological variables that had been
significantly different between PTSD and control subjects (SICI,
ICF, and SAI) were entered into a stepwise multiple regression
analysis. The significance level was set at p ? .05.
The two groups of subjects did not differ in education (U ?
120.00; p ? .211; n ? 36), gender [?2(1) ? .13; p ? .709], number
of smokers [?2(1) ? .024; p ? .877], or in the number of
postmenopausal women and menstrual cycle phase at time of
testing [?2(2) ? .486; p ? .784]. Because age was close to the
level of significance [t(34) ? ?2.02; p ? .051], it was used as
covariate for all further analyses. Information regarding illness
duration, PTSD severity measured with the CAPS, and depressive
and anxiety symptoms measured, respectively, with HAM-D and
HAM-A are listed in Table 1.
Thirteen of the PTSD subjects were victims of terrorist attacks,
3 subjects were victims of a car accident, 3 subjects had experi-
enced the sudden death of a family member, and only 1 person
was victim of physical abuse. All PTSD subjects had experienced
the traumatic event during adulthood. According to the total
score of the CAPS, all subjects suffered from moderate to severe
Repeated-measures multivariate ANCOVA with hemisphere
(left vs. right) as the repeated factor, group and gender as the
between-subject factors, age as covariate, and the TMS variables
(RMT, SICI, ICF, CSP, SAI) as the within-subjects factors showed
a significant difference between groups [F(5,19) ? 5.26; p ?
.003], no significant main effect for age [F(5,19) ? .64; p ? .668],
and no significant main effect for hemisphere (left vs. right)
[F(5,19) ? 1.18; p ? .352]. No interactions between age and
hemisphere [F(5,19) ? 1.47; p ? .243] and diagnosis and
hemisphere [F(5,19) ? 2.46; p ? .070] were found.
Group analysis also highlighted significant differences be-
tween the two groups in SICI [F(1,25) ? 6.66; p ? .016], ICF
[F(1,25) ? 16.28; p ? .001], and SAI [F(1,25) ? 10.64; p ? .003]
but not in the CSP [F(1,25) ? 3.30; p ? .081] and RMT [F(1,25) ?
2.22; p ? .149].
Post hoc univariate ANCOVAs with diagnosis as the between
factor and age as covariate showed that PTSD subjects had
reduced SICI in the left hemisphere [F(1,35) ? 9.79; p ? .004] but
not in the right hemisphere [F(1,35) ? 5.38; p ? .027] and
increased ICF in the right hemisphere [F(1,35) ? 12.15; p ? .001]
but not in the left one [F(1,35) ? 3.63; p ? .065]. In approxi-
mately 50% of PTSD patients, the SICI was reversed to facilitation
in both hemispheres, whereas this occurred only in one control
subject in the right hemisphere (Figure 1). No significant differ-
ences between patients and control subjects were found for CSP
in either left [F(1,35) ? 3.01; p ? .092] or right hemisphere
[F(1,35) ? 1.79; p ? .189] or for RMT in either left [F(1,35) ? .48;
p ? .487] or right hemisphere [F(1,35) ? .93; p ? .342]. Finally,
no significant difference was found between the two groups for
SAI in left hemisphere [F(1,27) ? 4.30; p ? .049], whereas in the
right hemisphere the difference between the two groups for the
SAI was very close to the significance level [F(1,27) ? 7.92; p ?
.009] (Figure 1, Table 2).
Relationships Between Clinical and TMS Variables
A stepwise multiple regression analysis between clinical
(illness duration, CAPS subscores, HAM-A and HAM-D) and TMS
variables (SICI, ICF, and SAI) showed a significant association
between illness duration and impaired SICI in the right hemi-
sphere (?, .445; p ? .049). The association between illness
duration and SAI in the right hemisphere was also very close to
the significance level (?, .517; p ? .053). The score of avoidance
subscale of CAPS was associated with SAI in the right hemisphere
(?, ?.937; p ? .016) and with increased ICF (?, .534; p ? .028),
also in the right hemisphere. The HAM-A and HAM-D scores
were not associated with either left or right SICI, ICF, or SAI.
Whereas it is generally accepted that the clinical/diagnostic
aid of a single TMS measure of cortical excitability is negligible
(14,38), the investigation of a wider set of TMS variables is
delineating patterns of neurophysiological changes both in psy-
56 BIOL PSYCHIATRY 2009;66:54–61
S. Rossi et al.
chiatric and neurological disorders, as depression (34,35), obses-
sive-compulsive disorders (OCD) (39), Gilles de la Tourette
(GTS) syndrome (40), Parkinson’s syndrome (41), and stroke
(42,43), as well as in subjects with distinctive personality traits
(44) or along their physiological aging (45).
At a group level, the current sample of drug-naive PTSD
patients showed bilateral SICI and SAI impairment, increased
right ICF, and normal CSP and RMT. After multiple comparisons,
the left-sided SICI impairment and the right-sided ICF increase
remained overtly significant, whereas the impairment of SAI in
the right hemisphere but not in the left one remained very close
to the significance level (see Table 2). Such a pattern of cortical
excitability dysfunction clearly differs, in terms of putative in-
volved neurotransmitter systems and laterality, from that of
drug-naive depressed patients (bilateral SICI and CSP impair-
ment, normal ICF) (34). Moreover, it still differs with that of
patients belonging to other anxiety spectrum disorders as GTS
with comorbid ADHD (increased RMT, impaired SICI and SAI,
increased ICF, only the left hemisphere tested , or OCD
[reduced RMT, impaired SICI, only left hemisphere tested] ).
Hence, whereas impaired SICI seems to be a common trait for
anxiety spectrum and depressive disorders, widening the avail-
able measures of cortical excitability might, theoretically, offer
more distinctive clues. However, it is difficult to determine the
level of specificity of these findings, due to lack of comparative
studies addressing the same TMS variables in both hemispheres
in patients belonging to the spectrum of anxiety disorders.
Alteration of Each TMS Variable: Physiological Significance
We found—as an extension of a previous study that disclosed
impaired SICI in the left hemisphere of PTSD patients, pointing to
a deficient GABAA-ergic inhibition even when symptoms are
mild (8)—a bilateral reverse of intracortical inhibition into facil-
itation in approximately 50% of the current sample of patients, a
finding likely due to both the longer illness duration and the
more severe clinical symptoms. This might suggest a more severe
and widespread loss of inhibition along disease progression,
particularly evident in the right hemisphere, as also suggested by
the significant association between SICI impairment and illness
duration. However, it is possible that a more severe underlying
disturbance in the balance of inhibition/excitation in these
patients leads to more severe symptoms or greater chronicity.
It might be argued that impaired SICI does not necessarily
reflect exclusive GABAAdysfunction, because at ISIs of 2–4 msec
overlapping short-term intracortical facilitation phenomena
(SICF) might take place (46). However, the fact that an overtly
Table 1. Demographic and Clinical Characteristics of Healthy Subjects and Subjects with PTSD
(n ? 16)PTSD (n ? 20)Analysis
Age (Yrs)40.8 13.950.514.6
Education (Yrs)14.93.613.33.3 120.0036.211
Menstrual Cycle Phase
Luteal (according to the last
Duration of Illness (Yrs)
Type of Trauma, n
Sudden death of a family
CAPS total score
HAM-D total score
HAM-A total score
PTSD, posttraumatic stress disorder; CAPS, Clinician Administered Posttraumatic Stress Disorder Scale; HAM-D,
Hamilton Depression Rating Scale; HAM-A, Hamilton Anxiety Rating Scale.
aUnpaired t test.
dNumber of cigarettes/day 15 ? 3.
S. Rossi et al.
BIOL PSYCHIATRY 2009;66:54–61 57
subthreshold (20% ? individual RMT) conditioning stimulus
would recruit a different set of cortical interneurons only in a
portion of PTSD patients but neither in the remaining patients
nor in control subjects except one (see Figure 1, SICI) would
imply a peculiar reactivity of these neural pools due to the
underlying disease and that a bilateral decrease of the net
inhibition resulting from the summation of SICI and SICF is
taking place, whatever the mechanism.
Impaired SAI was evident at a group level and was only close
to the significance level in the right hemisphere (p ? .009) after
Bonferroni adjustment. However, it was significantly associated
with the avoidance score. The SAI is likely to reflect impaired
presynaptic GABAAmechanisms (26) rather than cholinergic
dysfunction (24,25), because evidences of cholinergic dysfunc-
tion in PTSD, postulated in animal models (47), are not yet
confirmed in humans. However, due to the lack of correlation
between SICI and SAI impairment in the right hemisphere, it is
unlikely that they are both modulated by the same GABAA-ergic
The RMT and CSP did not differ in the two groups or between
right and left hemisphere in PTSD patients. The RMT is increased
by administration of drugs blocking voltage-gated sodium chan-
nels (12–14) but is insensitive to drugs acting as GABA- (12) or
NMDA-glutamate modulators (22,23). According to these phar-
macological effects, GABA/glutamate imbalance in PTSD did not
More intriguing is that no difference in CSP, a putative
GABAB-mediated phenomenon (17,18), has been found. This
would signify that widespread loss of inhibition in PTSD
patients, as reflected here by impaired SICI (and SAI) and
possibly by reduction of plasmatic GABA levels (48), is mainly
due to a defective GABAA-mediated central transmission, in
and in the 20 posttraumatic stress disorder (PTSD) patients. From the top/left: short-latency intracortical inhibition (SICI), intracortical facilitation (ICF),
short-latency afferent inhibition (SAI) (15 control subjects and 14 patients), cortical silent period (CSP), and resting motor threshold (MT). Statistics are in the
text. MEP, motor-evoked potential.
58 BIOL PSYCHIATRY 2009;66:54–61
S. Rossi et al.
keeping with neuroimaging investigations (5). This is also
supported by decreased cerebrospinal fluid allopregnanolone
levels in women with PTSD (49) and by the low post-trauma
GABA plasma levels as a predictive factor in the development
of acute PTSD (50).
Interestingly, recent therapeutic attempts for PTSD have in-
cluded tiagabine as a potential GABAA–mimetic agent in an open
study (51), although a larger placebo-controlled investigation did
not confirm its clinical utility in controlling PTSD symptoms (52).
Exaggerated ICF in the right hemisphere and its association
with symptoms of avoidance are discussed in the following
Interhemispheric Imbalance of Cortical Excitability
Inter-hemispheric imbalance in PTSD patients, with right-
sided functional prevalence, is not a new finding and seems to be
consistent with the leading role of the right hemisphere in
processing emotions, integrating sensory modalities, and retriev-
ing memories (53–56). Electroencephalographic (EEG) activation
in right anterior and posterior regions of PTSD patients who had
motor vehicle accidents, lacking in the resting state (57,58),
became apparent during processing of specific traumatic stres-
sors (57). These EEG findings fit nicely with the relative increase
of regional cerebral blood flow in the right hemisphere of PTSD
patients exposed to trauma recall (53,59) and with the “reactivity”
of the right hemisphere to repetitive TMS procedures, which
transiently alleviate PTSD symptoms (60).
The current pattern of cortical excitability dysfunctions con-
firms the widespread impaired inhibition and suggests a possible
role of the relative prevalence of glutamatergic tone in the right
hemisphere, as indexed by exaggerated ICF, in the pathophysi-
ology of PTSD. The fact that symptoms of PTSD might be partly
related to a glutamatergic dysfunction is an emerging hypothesis
(9,10) that seems supported by the clinical efficacy of such
antiglutamatergic drugs as lamotrigine (61) on PTSD symptoms;
however, this hypothesis cannot be proven by neuroimaging
methods, because reliable glutamate tracers are still lacking (27).
Notably, GABA/glutamatergic unbalance in animal models of
PTSD is thought to increase nitric oxide synthase, which might
have a role in stress-related hippocampal degeneration (2). This
mechanism might underlie reduction of hippocampal volume in
adult PTSD patients (62). Indeed, PTSD patients who underwent
a parallel morphometric magnetic resonance imaging study at
our center had smaller than normal hippocampal and overall
gray matter volume (63). A follow-up study is currently under
way that is aimed at verifying whether chronic serotoninergic
therapy, besides restoring hippocampal volume (64,65), can also
re-establish a normal pattern of cortical excitability.
Relationships with Symptoms
In the current PTSD sample, impaired SICI in the right
hemisphere, reflecting an overall loss of GABAA-mediated inhi-
bition, correlated with illness duration but not with anxiety.
There was also an association between symptoms of avoidance,
increased ICF, and impaired SAI, again in the right hemisphere.
This finding could be consistent with the notion that the longer
the disease, the more likely is the occurrence of lasting functional
or structural changes, even in brain regions such as the motor
cortex that are outside although functionally connected with the
anatomical circuitries that are usually found dysfunctional in
PTSD, including anterior cingulated cortex, amygdala, limbic and
paralimbic structures, Broca’s area, and other neocortical regions
(66,67). This is scarcely surprising, because TMS procedures are
particularly sensitive in disclosing early subclinical changes of
the motor system, even for diseases such as Alzheimer’s demen-
tia in which the motor impairment can eventually occur only in
the very final stage (68).
Significance of Findings and Limitations of the Study
The described relationships between cortical excitability dys-
functions and the pathophysiology of PTSD symptoms are
theoretical. A causative support will eventually be confirmed by
prospective studies demonstrating that, when symptoms are
reduced by a successful treatment (either pharmacological or
cognitive), cortical excitability is modified in the same direction
and vice versa (i.e., further degradation of neurophysiological
indexes by exposing PTSD patients to “stressors”).
Because most persons exposed to trauma do not develop
PTSD, it has been suggested that a specific phenotype could be
associated with a failure to recover from the normal conse-
quences of trauma (69). Therefore, current neurophysiological
abnormalities might be associated with PTSD symptoms—as
suggested by the association of right-sided TMS variables with
the illness duration and avoidance scores—or might be related to
a certain preexisting endophenotype. Only the use of a trauma-
exposed control population not developing PTSD might help to
clarify this issue.
Considering that severe PTSD is often resistant to pharmaco-
logical and cognitive therapies, the pattern of alteration of
cortical excitability might offer the anatomic-functional rationale
for choosing the best target to be used in future neuromodulatory
treatment strategies, both noninvasive (like repetitive TMS) or
minimally invasive (like chronic epidural stimulation), in analogy
with what is being done with other psychiatric or chronic
neurological diseases associated with cortical excitability dys-
We thank Patrizio Pasqualetti for statistical advice and
Giovanni Filippone and Federica Felici for experimental help.
Table 2. Comparison of TMS Variables Between Control Subjects and
(n ? 16)
(n ? 20)
MeanSD Mean SD
TMS, transcranial magnetic stimulation; PTSD, posttraumatic stress dis-
order; SICI, short latency intracortical inhibition; ICF, short latency intracor-
tion; RMT, resting motor threshold.
aUnivariate analysis of variance.
bDenotes significance at Bonferroni adjusted ? ? .0083 for contrasts.
dControl subjects n ? 16; PTSD n ? 12.
S. Rossi et al.
BIOL PSYCHIATRY 2009;66:54–61 59
All authors report no biomedical financial interests or poten-
tial conflicts of interest.
ual of Mental Disorders, 4th ed. Washington, DC: American Psychiatric
2. Harvey BH, Oosthuizen F, Brand L, Wegener G, Stein DJ (2004): Stress-
3. Matsumoto K, Puia G, Dong E, Pinna G (2007): GABAAreceptor neuro-
transmission dysfunction in a mouse model of social isolation-induced
SSRIs in mood and anxiety disorders. Stress 10:3–12.
4. Pibiri F, Nelson M, Guidotti A, Costa E, Pinna G (2008): Decreased corti-
5. Geuze E, van Berckel BN, Lammertsma AA, Boellaard R, de Kloet CS,
Vermetten E, et al. (2007): Reduced GABAAbenzodiazepine receptor
binding in veterans with post-traumatic stress disorder. Mol Psychiatry
6. Bremner JD, Innis RB, Southwick SM, Staib L, Zoghbi S, Charney DS
(2000): Decreased benzodiazepine receptor binding in prefrontal cor-
tex in combat-related posttraumatic stress disorder. Am J Psychiatry
7. Fujita M, Southwick SM, Denucci CC, Zoghbi SS, Dillon MS, Baldwin RM,
with posttraumatic stress disorder. Biol Psychiatry 56:95–100.
et al. (2005): Cortical hyperexcitability in post traumatic stress disor-
ders secondary to minor accidental head trauma: A neurophysio-
logic study. J Psychiatry Neurosci 30:127–132.
9. Nutt DJ (2000): The psychobiology of posttraumatic stress disorders.
matic stress disorder. CNS Spectr 13:585–591.
et al. (1994): Non-invasive electrical and magnetic stimulation of the
brain, spinal cord and roots: Basic principles and procedures for
routine clinical application. Report of an IFCN Committee. Electroen-
cephalogr Clin Neurophysiol 91:79–92.
12. Ziemann U, Lonnecker S, Steinhoff BJ, Paulus W (1996): The effect of
lorazepam on the motor cortical excitability in man. Exp Brain Res 109:
13. Ziemann U (2004): TMS and drugs. Clin Neurophysiol 115:1717–1729.
14. Chen R, Cros D, Currà A, Di Lazzaro V, Lefaucheur JP, Magistris MR, et al.
(2008): The clinical diagnostic utility of transcranial magnetic stimula-
tion: Report of an IFCN Committee. Clin Neurophysiol 119:504–532.
lation: The silent period after the motor evoked-potential. Neurology
16. Inghilleri M, Berardelli A, Marchetti P, Manfredi M (1996): Effects of
diazepam, baclofen and thiopental on the silent period evoked by
transcranial magnetic stimulation in humans. Exp Brain Res 109:467–
cal baclofen infusions induced a marked increase of the transcranially
evoked silent period in a patient with generalized dystonia. Muscle
18. Werhahn KJ, Kunesch E, Noachtar S, Benecke R, Classen J (1999): Differ-
ential effects on motorcortical inhibition induced by blockade of GABA
uptake in humans. J Physiol Lond 517:591–597.
19. Di Lazzaro V, Oliviero A, Meglio M, Cioni B, Tamburrini G, Tonali P, et al.
ity of the human motor cortex. Clin Neurophysiol 111:794–799.
20. Kujirai T, Caramia MD, Rothwell JC, Day BL, Thompson PD, Ferbert A,
et al. (1993): Corticocortical inhibition in human motor cortex.
J Physiol 471:501–519.
21. Ilic TV, Meintzschel F, Cleff U, Ruge D, Kessler KR, Ziemann U (2002):
Short interval paired-pulse inhibition and facilitation of human motor
22. Ziemann U, Chen R, Cohen LG, Hallett M (1998): Dextromethorphan
decreases the excitability of the human motor cortex. Neurology 51:
23. Liepert J, Schwenkreis P, Tegenthoff M, Malin JP (1997): The glutamate
antagonist riluzole suppresses intracortical facilitation. J Neural Transm
(2000): Muscarinic receptor blockade has differential effects on the
25. Di Lazzaro V, Oliviero A, Pilato F, Saturno E, Dileone M, Marra C, et al.
(2005): Neurophysiological predictors of long term response to AChE
Segregating two inhibitory circuits in human motor cortex at the level
of GABAAreceptor subtypes: A TMS study. Clin Neurophysiol 118:2207–
27. Yu M (2007): Recent developments of the PET imaging agents for
metabotropic glutamate receptor subtype 5. Curr Top Med Chem
28. First MB, Spitzer RL, Gibbon M, Williams JB (1996): Structured Clinical
29. First MB, Spitzer RL, Gibbon M, Williams JB (1994): Structured Clinical
Interview For Axis II DSM IV Disorders (SCID-II/P). New York: Biometrics
Research Department, New York State Psychiatric Institute.
30. First MB, Spitzer RL, Gibbon M, Williams JB (1995): Structured Clinical
Interview for Axis I DSM IV Disorders, Non-Patient Edition (SCID-I/NP, ver-
sion 2.0). New York: Biometrics Research Department, New York State
31. Blake DD, Weathers FW, Nagy LM, Kaloupek DG, Gusman FD, Charney
DS, et al. (1995): The development of a clinician-administered PTSD
32. Hamilton M (1960): A rating scale for depression. J Neurol Neurosurg
depression. Biol Psychiatry 59:395–400.
35. Leafucheur JP, Lucas B, Andraud F, Hogrel JY, Belliver F, Del Cul A, et al.
sensory input from the hand. J Physiol Lond 523:503–513.
38. Rossini PM, Rossi S (2007): Transcranial magnetic stimulation. Diagnos-
tic, therapeutic and research potential. Neurology 68:484–488.
JC, et al. (2000): Altered cortical excitability in obsessive-compulsive
disorder. Neurology 54:142–147.
41. Leafucheur JP (2005): Motor cortex dysfunction revealed by cortical
treatment and cortical stimulation. Clin Neurophysiol 116:244–253.
43. Swayne OB, Rothwell JC, Ward NS, Greenwood RJ (2008): Stages of
motor output reorganization after hemispheric stroke suggested by
longitudinal studies of cortical physiology. Cereb Cortex 18:1909–1922.
44. Wassermann EM, Greenberg BD, Nguyen MB, Murphy DL (2001): Motor
cortex excitability correlates with an anxiety-related personality trait.
60 BIOL PSYCHIATRY 2009;66:54–61
S. Rossi et al.
of aging brain: From normal aging to neurodegeneration. Prog Neuro-
46. Peurala SH, Müller-Dahlhaus JF, Arai N, Ziemann U (2008): Interference
of short-interval intracortical inhibition (SICI) and short-interval intra-
cortical facilitation (SICF). Clin Neurophysiol 119:2291–2297.
long-lasting changes in cholinergic gene expression. Nature 393:373–
48. Vaiva G, Boss V, Ducrocq F, Fontaine M, Devos P, Brunet A, et al. (2006):
Relationship between posttrauma GABA plasma levels and PTSD at
49. Rasmusson AM, Pinna G, Paliwal P, Weisman D, Gottschalk C, Charney
in women with PTSD. Biol Psychiatry 60:704–713.
opment of acute posttraumatic stress disorder. Biol Psychiatry 55:250–
51. Connor KM, Davidson JR, Weisler RH, Zhang W, Abraham K (2006):
Tiagabine for posttraumatic stress disorder: Effects of open-label
and double-blind discontinuation treatment. Psychopharmacology
52. Davidson JR, Brady K, Mellman TA, Stein MB, Pollack MH (2007): The
efficacy and tolerability of tiagabine in adult patients with post-trau-
matic stress disorder. J Clin Psychopharmacol 27:85–88.
53. Rauch SL, van der Kolk BA, Fisler RE, Alpert NM, Orr SP, Savage CR, et al.
(1996): A symptom provocation study using positron emission tomog-
raphy and script driven imagery. Arch Gen Psychiatry 53:380–387.
55. Rossi S, Cappa SF, Babiloni C, Pasqualetti P, Miniussi C, Carducci F, et al.
(2001): Prefrontal cortex in long-term memory: An “interference” ap-
proach using magnetic stimulation. Nat Neurosci 4:948–952.
56. Rossi S, Cappa SF, Ulivelli M, De Capua A, Bartalini S, Rossini PM (2006):
rTMS for PTSD: Induced merciful oblivion or elimination of abnormal
hypermnesia? Behav Neurol 17:195–199.
electrical activity in posttraumatic stress disorder after motor vehicle
58. Shankman SA, Silverstein SM, Williams LM, Hopkinson PJ, Kemp AH,
Felmingham KL, et al. (2008): Resting electroencephalogram asymme-
try and posttraumatic stress disorder. J Trauma Stress 21:190–198.
exposed to assaultive and non-assaultive trauma and developing or not
60. Cohen H, Kaplan Z, Kotler M, Kouperman I, Moisa R, Grisaru N (2004):
Repetitive transcranial magnetic stimulation of the right dorsolateral
prefrontal cortex in posttraumatic stress disorder: A double-blind, pla-
Connor KM, et al. (1999): A preliminary study of lamotrigine for the
treatment of posttraumatic stress disorder. Biol Psychiatry 45:1226–
62. Woon FL, Hedges DW (2008): Hippocampal and amygdala volumes in
children and adults with childhood maltreatment-related posttrau-
matic stress disorder: A meta-analysis. Hippocampus 18:729–736.
(2008): Magnetic resonance imaging volumes of the hippocampus in
64. Vermetten E, Vythilingam M, Southwick SM, Charney DS, Bremner JD
(2003): Long-term treatment with paroxetine increases verbal declara-
(2007): Changes in hippocampal volume in patients with post-trau-
matic stress disorder after sertraline treatment. J Clin Psychopharmacol
posttraumatic stress disorder with neuroimaging. J Clin Psychiatry 62:
67. Bremmer JD, Elzinga B, Schmahal C, Vermetten E (2008): Structural and
functional plasticity of the human brain in posttraumatic stress disor-
68. Ferreri F, Pauri F, Pasqualetti P, Fini R, Dal Forno G, Rossini PM (2003):
stimulation study. Ann Neurol 53:102–108.
69. Yehuda R, LeDoux J (2007): Response variation following trauma: A
translational neuroscience approach to understanding PTSD. Neuron
70. Leafucheur JP (2008): Principles of therapeutic use of transcranial and
epidural cortical stimulation. Clin Neurophysiol 119:2179–2184.
S. Rossi et al.
BIOL PSYCHIATRY 2009;66:54–61 61