consciousness with high-density EEG in altered states of ... · with the most recent theories of consciousness and this study concludes with the perspectives of what TMS-EEG can offer.

Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=ibij20

Download by: [University of Liege] Date: 20 March 2017, At: 02:27

Brain Injury

ISSN: 0269-9052 (Print) 1362-301X (Online) Journal homepage: http://www.tandfonline.com/loi/ibij20

Transcranial magnetic stimulation combinedwith high-density EEG in altered states ofconsciousness

Martino Napolitani, Olivier Bodart, Paola Canali, Francesca Seregni,Adenauer Casali, Steven Laureys, Mario Rosanova, Marcello Massimini &Olivia Gosseries

To cite this article: Martino Napolitani, Olivier Bodart, Paola Canali, Francesca Seregni,Adenauer Casali, Steven Laureys, Mario Rosanova, Marcello Massimini & Olivia Gosseries(2014) Transcranial magnetic stimulation combined with high-density EEG in altered states ofconsciousness, Brain Injury, 28:9, 1180-1189, DOI: 10.3109/02699052.2014.920524

To link to this article: http://dx.doi.org/10.3109/02699052.2014.920524

Published online: 06 Aug 2014.

Submit your article to this journal

Article views: 425

View related articles

View Crossmark data

Citing articles: 1 View citing articles

http://informahealthcare.com/bijISSN: 0269-9052 (print), 1362-301X (electronic)

Brain Inj, 2014; 28(9): 1180–1189! 2014 Informa UK Ltd. DOI: 10.3109/02699052.2014.920524

Transcranial magnetic stimulation combined with high-density EEG inaltered states of consciousness

Martino Napolitani1*, Olivier Bodart2*, Paola Canali1, Francesca Seregni1, Adenauer Casali1, Steven Laureys2,Mario Rosanova1, Marcello Massimini1,3, & Olivia Gosseries2,4,5

1Department of Biomedical and Clinical Sciences ‘Luigi Sacco’, University of Milan, Milan, Italy, 2Coma Science Group, Cyclotron Research Centre

and Neurology Department, University and University Hospital of Liege, Liege, Belgium, 3Istituto Di Ricovero e Cura a Carattere Scientifico,

Fondazione Don Carlo Gnocchi, Milan, Italy, 4Center for Sleep and Consciousness, Department of Psychiatry, and 5Postle Laboratory, Department of

Psychology and Psychiatry, University of Wisconsin, Madison, WI, USA

Abstract

Background: This review discusses the advantages of transcranial magnetic stimulationcombined with high-density electroencephalography (TMS-hdEEG) over other currenttechniques of brain imaging.Methods and results: Its application was reviewed, focusing particularly on disorders ofconsciousness, in the perspective of recent theories of consciousness. Assessment ofnon-communicative patients with disorders of consciousness remains a clinical challenge andobjective measures of the level of consciousness are still needed. Current theories suggest thata key requirement for consciousness is the brain’s capacity to rapidly integrate informationacross different specialized cortical areas. TMS-EEG allows the stimulation of any given corticalarea and the recording of the immediate electrical cortical response. This technique hasrecently been successfully employed to measure changes in brain complexity underphysiological, pharmacological and pathological conditions.Conclusions: This suggests that TMS-EEG is a reliable tool to discriminate between consciousand unconscious patients at the single subject level. Future works are needed to validate andimplement this technique as a clinical tool.

Keywords

Consciousness, effective connectivity,electroencephalography, minimallyconscious state, transcranial magneticstimulation, unresponsive wakefulnesssyndrome, vegetative state

History

Received 9 August 2013Revised 28 February 2014Accepted 12 March 2014Published online 25 July 2014

Introduction

Despite considerable progress, patients with disorders of

consciousness (DOC) such as those in a vegetative state

(recently renamed unresponsive wakefulness syndrome, UWS

[1]) or in a minimally conscious state (MCS [2]) still pose a

challenge to both medical teams and families. After falling

into a coma, patients with UWS recover arousal but not

consciousness, whereas patients in MCS show reproducible

but fluctuating signs of consciousness (e.g. responses to

command, visual pursuit). The diagnosis of consciousness is

mainly based on behavioural evaluation, which can give rise

to misdiagnoses [3]. This problem is only partially resolved

by the introduction of standardized dedicated scales such as

the Coma Recovery Scale–Revised (CRS-R [4]). Numerous

neuroimaging and electrophysiological paradigms have been

developed in recent years, leading to a better understanding of

the neural correlates of consciousness, thereby allowing for

the detection of differences between patient groups. However,

at the single subject level, all these advancements lack

sensitivity and/or specificity that are required in a clinical

setting. In parallel, new theories of consciousness suggest that

a perturbational approach would bring relevant information

about the brain’s capacity for consciousness [5]. Transcranial

magnetic stimulation (TMS) coupled with high-density elec-

troencephalography (EEG) [6] is a technique that allows the

stimulation of any given cortical area and the recording of the

immediate electrical cortical response. TMS-EEG has been

successfully employed to discriminate between conscious and

unconscious patients under physiological, pharmacological

and pathological conditions.

This review will discuss the recent findings obtained with

TMS-EEG in various states of consciousness, with a particu-

lar focus on DOC. These results will be explored along

with the most recent theories of consciousness and this

study concludes with the perspectives of what TMS-EEG can

offer.

Evaluation of consciousness

Current evaluations of the residual level of consciousness rely

on behavioural assessments. These evaluations have an

intrinsic limitation: they need the subject to produce a

*contributed equally.

Correspondence: Olivia Gosseries, Coma Science Group, CyclotronResearch Centre and Neurology Department, University and UniversityHospital of Liege, Sart-Tilman B30, 4000 Liege, Belgium.Tel: +32 4 366 23 16. Fax: +32 4 366 29 46. E-mail: [email protected]

motor output, either to a visual, tactile or auditory stimulus.

They also usually require patients to understand the task and

be willing to collaborate [7, 8]. This behavioural approach

leaves the physician unaware of a possible impairment of

motor and/or afferent projections, likely to be present in

patients with severe brain injury [8]. In order to bypass these

limitations, neuroscientists have developed different tech-

niques to assess consciousness without relying on motor

output. Functional magnetic resonance imaging (fMRI) using

mental imagery paradigms can detect consciousness via

wilful activation of specific brain areas in some patients with

DOC [9–12]. However, the lack of response should not be

interpreted as an absence of consciousness. In fact, many

patients in MCS, despite showing some signs of conscious-

ness at the bedside, do not show task-related brain activation,

which might be due to a lack of comprehension, attention,

vigilance or motivation [7, 11]. Besides active task proced-

ures, results from passive paradigms using various external

stimulations have shown differences at the group level

between patients in MCS and UWS, but not at the individual

level [13–15]. Ultimately, the brain can be studied at rest,

without performing any task and without receiving any

stimulation. Recent fMRI studies show that functional

connectivity in the default mode network (which is involved

in self-related processes such as inner speech and mind

wandering) correlates with the level of consciousness in

patients with DOC [16, 17]. However, some have questioned

the validity of such indirect tools to assume one’s conscious-

ness [18]. While functional neuroimaging has an excellent

spatial resolution, it lacks temporal resolution because it

cannot record events shorter than several seconds. This can be

problematic as the time scale for capturing consciousness is

thought to be in the order of hundreds of milliseconds [19].

Moreover, functional connectivity does not provide any causal

interpretation.

Conversely, electrophysiological measurements can detect

neuronal activities in a timescale of milliseconds. Active

paradigms have also been employed with EEG, allowing the

detection of wilful activities following specific instruction in

several patients otherwise considered as unconscious [20–22].

Many sophisticated EEG analyses have been developed, such

as the bispectral index [23, 24], spectral and entropy analyses

[25, 26], neuronal complexity [27], causal density [28] and

Granger causality [29], but these procedures solely deal with

the background neuronal activity and, thus, lack the deter-

ministic value of a perturbational approach.

Although neuroimaging and EEG techniques are able to

bypass some of the behavioural evaluations’ limitations and

some show promising results at the group level, none can

presently be used at the single subject level. Some authors

suggest that TMS-EEG is a good candidate to achieve

accurate evaluation of consciousness at the single subject

level [5, 30].

TMS-EEG techniques

One way to perturb the brain non-invasively is to force

neurons in a small area of the cortex to depolarize, with the

expectation that this excitation would propagate to other brain

regions. TMS allows researchers to do so by means of

electromagnetic induction [5]. These magnetic pulses induce

secondary ionic currents in the brain that penetrate the

membranes of the neurons, resulting in an action potential or

excitatory/inhibitory post-synaptic potential. As the magnetic

field falls off rapidly with distance from the coil [5, 31], it

only directly activates neural elements superficially (that is,

on the surface of the cortex) or in the white matter underneath

the stimulation site, limiting the number of brain’s structures

one can reach. The amount of neurons and the layers that are

directly triggered by the TMS remain undetermined.

However, experimental [32] and modelling [33] studies

strongly suggest that axons, rather than cell bodies, are

most likely the targets of the stimulation. Evidently, axons

have the lowest threshold for activation to the brief electrical

currents induced by TMS. The capacity of TMS to depolarize

neurons depends on the ‘activating function’, which causes a

sufficient transmembrane current to flow and depolarize the

membrane. Stimulation will take place at the point where the

spatial derivative of the induced electric field is maximal.

From the site of cortical stimulation, the neuronal signal then

propagates along intra- and inter-hemispheric association

fibres to other cortical areas and to deeper neural structures.

However, the neural signal also, via projecting fibres,

propagates to sub-cortical structures and sometimes to the

spinal cord. The ‘effective connectivity’ measures the causal

interaction between the differently specialized brain

areas, which is different from ‘functional connectivity’

(that measures temporal correlation) and from ‘structural

connectivity’ (that assesses the anatomical links between

brain regions) [34].

While TMS systems have been available for a long time,

EEG amplifiers have not been compatible and recording

artefact-free response was impossible. EEG amplifiers have

then been equipped with sample-and-hold circuits to prevent

the recording of the powerful TMS-related artefact. This

allowed recording the TMS evoked potentials response with

milliseconds time scale, which reliably reflets the state of

excitability of underlying cortical circuits [6, 35–38]. The

EEG cap is often composed of 60 flat open ring carbon

electrodes specially designed to further decrease TMS related

artefacts, along with one reference, one ground and two

electrooculogram electrodes (Figure 1). Muscles artefacts are

also reduced to their minimum by stimulating central scalp

regions, avoiding the temporal muscles masses [39]. In order

to avoid the contamination of the responses by auditory

evoked EEG signal, noise masking is applied through

earphone playing white noise in the frequency band of the

TMS coil ‘click’ [40]. To further standardize stimulation

parameters, ensuring reproducibility and accuracy over cor-

tical areas and across subjects, the experimental set-up should

be equipped with a stereotactic neuronavigation system

(Figure 1) [41]. The neuronavigation system is used with a

structural image of the brain of the subject and locates the

relative position of the subject’s head and the TMS coil.

Moreover, the system may take into account the relative

scalp-to-cortex distance to calculate the electric field induced

by TMS on the cortical surface. This ensures that the effective

stimulation intensity is the same across subjects and stimu-

lation sites. Therefore, the combination of neuronavigation

system, TMS and compatible EEG allows the acquisition of

DOI: 10.3109/02699052.2014.920524 TMS-EEG in disorders of consciousness 1181

cortical evoked potentials that are triggered and recorded

from cortical structures. The amplitude, spreading and

morphology of the evoked potentials are the characteristics

considered in order to investigate the brain’s excitability and

connectivity.

Regarding the analysis, the EEG data sets are

pre-processed to remove any trial contaminated with eye

movements, blinks, muscle or movement artefact, before a

source-modelling algorithm is applied. Significant signals

from EEG sensors are transposed to cortical sources, accord-

ing to the time sequence and the topography of the registered

evoked response. This transposition can be performed on the

subject’s own brain imagery or on a standard model [42].

Casali et al. [42] developed a method to further analyse

TMS-EEG data with three synthetic indices: significant

current density (SCD), significant current scattering (SCS)

and phase locking (PL). SCD sums up the amplitude of

all significant currents induced by TMS, SCS measures the

average distance of significantly activated sources from

the stimulated region, while PL reflects the ability to re-set

the ongoing phase of cortical oscillations; these indices,

therefore, help to assess different aspects of brain respon-

siveness. Table I summarizes the advantages and disadvan-

tages of the TMS-EEG technique. Considering this evidence,

this technique seems to be a promising tool to assess brain

functions in both healthy subjects and pathological condi-

tions. Specifically, it can be a useful method in the assessment

of non-communicative patients, as discussed in the next

sections.

Experimental recordings in healthy awake subjects

Some recent studies employed TMS-EEG to measure cortical

excitability (i.e. the amplitude of the initial response to TMS,

which can also be assessed by SCD) and effective connect-

ivity (i.e. the causal interaction between the stimulated area

and the subsequent activated cortical regions, which can be

assessed by SCS) in healthy awake subjects. The responses

were recorded during rest and various cognitive tasks such

as memory [43, 44], visual attention [45, 46] and motor

planning [47].

Massimini et al. [35, 48] were first to demonstrate that,

during wakefulness, TMS is followed by multiple low

amplitude fast EEG waves, associated with spatially and

temporally differentiated patterns of activation. Brain areas

also respond differently to a given stimulation, as shown by

Rosanova et al. [49], who delivered TMS on different brain

regions of healthy awake subjects. Regardless of the

Figure 1. Possible set-up, combining aneuronavigation system (a), the stimulationcoil with tracking elements (b), high-densityEEG net (c) and EEG amplifier (d). Theneuronavigation system is composed of 3Dbrain reconstruction (a1), an infrared trackingcamera (a2) and tracking googles (a3).

Table I. Advantages and disadvantages of TMS-EEG.

Advantages Disadvantages

� Bypass afferent sensory pathways.� Does not require functioning efferent pathways.� Does not require subject active participation.� Does not require language processing.� Highly reproducible within subject.� Can be use at the patient’s bedside.� Sensitive to changes in stimulation parameters.� Good temporal resolution.� Probes effective connectivity.� Discrimination between conscious and unconscious conditions.� Supported by recent theories of consciousness.

� Dependent on the subject cortical excitability, lowered in case of brainatrophy (it increases scalp-to-cortex distance) and with several drugs(including anti-epileptic).

� Requires stable state of wakefulness.� Limited spatial resolution.� Acute patients assessment limited by the presence of metallic implant,

external CSF drain or uncontrolled epilepsy.� Requires considerable logistic and subject preparation.� Source modelling possibly inaccurate in cases of extensive brain lesions or

scalp deformations.

CSF, Cerebrospinal fluid.

1182 M. Napolitani et al. Brain Inj, 2014; 28(9): 1180–1189

stimulation intensity, TMS evoked alpha band oscillations (8–

12 Hz) in the occipital cortex, beta band oscillations

(13–20 Hz) in the parietal cortex and fast beta/

gamma oscillations (21–50 Hz) in the frontal cortex.

TMS-EEG can, therefore, be used to directly investigate the

properties of thalamocortical circuits to produce natural

oscillations.

Huber et al. [50] showed that the cortical excitability

steadily increased in healthy subjects during the course of

prolonged wakefulness. Similarly, attention to motor tasks,

even without actual movement, causes an increase of local

cortical excitability, as assessed by TMS-EEG [47]. During

visual attention tasks, an increase of effective connectivity

has also been observed between the frontal eye field (the site

of stimulation) and posterior brain regions [46]. This effect

is thought to reflect a top-down control over these latter

regions. Another recent study showed that, during a spatial

working memory task, the delivery of TMS on a specific

brain area (i.e. the superior parietal lobule, an area known to

play a role in working memory) during the delay period also

induced a marked increase of both SCD and SCS, as

compared to a condition of rest (where subjects fixated

a static point) [43]. Moreover, Kundu et al. [44] have

recently demonstrated that a long-term training of a working

memory visual task resulted, once again, in an increase in

SCS in the task-related networks. Further evidence of task-

related network specific change of cortical excitability has

been provided in another recent study on face recognition,

where excitability was modulated in the medial prefrontal

cortex, but not in the premotor cortex [51].

In summary, these findings in healthy awake subjects

converge to the assumption that cognitive functions induce

increased cortical excitability and/or effective connectivity.

It is noted that alcohol intake reduces cortical excitability,

especially on the frontal and prefrontal areas [52].

Reduced brain excitability and decreased effective connect-

ivity have also been observed in various states of altered

consciousness.

Experimental recordings in altered states ofconsciousness

Physiological and pharmacological states

If awakened from an early non-rapid eye movement (NREM)

or slow-wave sleep, subjects report little or no conscious

content, while, if awakened from REM sleep, they are able

to report dreams, which can be as vivid as conscious

experience during wakefulness [53]. Different thalamocor-

tical networks responses have been recorded with TMS-EEG

during transition from wakefulness to NREM and REM

sleep. In NREM sleep, TMS triggers a larger positive–

negative low frequency wave. Depending on stimulation

parameters such as intensity, this response can stay local and

be transient or can be global and look like typical slow

waves [54]. Sleep-stage one induced an intermediate

response between wakefulness and NREM sleep and low

intensity stimulation evokes a typical slow wave with

disappearance of the latter components [35]. In contrast, in

REM sleep, the brain’s response to TMS is similar to the

one observed during wakefulness, with fast oscillatory

components, especially during the first 100–150 millisec-

onds post-stimulus [48] (Figure 2). These findings suggest

that TMS-EEG is able to monitor graded changes in the

level of consciousness during normal physiological changes.

Note also that, after a night of sleep, cortical excitability

decreases, as opposed to the increase observed after

prolonged wakefulness [50].

Pharmacological agents inducing a loss of consciousness,

such as midazolam at anaesthetic concentrations, have also

been investigated with TMS-EEG. Midazolam targets

GABA-A receptors exclusively, increasing inhibitory post-

synaptic currents and possibly causing its anaesthetic effect

[55]. In this condition, TMS evokes a stereotypical large

positive–negative wave, similar to the one evoked in NREM

sleep, which remains localized to the stimulation site

(Figure 2) [36]. Even if the mechanisms of action differ,

Figure 2. Typical response to TMS in different physiological, pharmacological and pathological conditions. For each condition, corticalresponses under the coil (bold black trace) and in seven other brain areas (lighter black traces) following right premotor stimulations (black lightning)are shown. (A) In NREM sleep, depending of stimulation intensity, TMS can trigger either a global or a local response, but both remain stereotypicaland slow. In REM sleep, the response is complex, spread in time and space, similar to the one of wakefulness. (B) General anaesthesia responses looklike NREM sleep responses. (C) In DOC, the response in patients in UWS is similar to the one in general anaesthesia and NREM sleep. In patients inMCS, the response is similar to the one observed in REM sleep and in wakefulness. TMS, Transcranial magnetic stimulation; REM, Rapid eyemovement; NREM, non-REM; UWS, Unresponsive wakefulness syndrome; MCS, Minimally conscious state. Adapted with permission from Sarassoet al. 2014 [76].

DOI: 10.3109/02699052.2014.920524 TMS-EEG in disorders of consciousness 1183

similar responses are expected with other pharmacological

inhibitory agents.

In summary, during wakefulness and REM sleep the brain

is able to sustain long-range and different activity patterns,

marked by a complex, diffuse and long-lasting evoked

response, whilst during NREM sleep and anaesthesia, when

consciousness fades, this ability is lost and, even if the

thalamocortical system remains active, it responds to a direct

TMS perturbation in a stereotypical and short lasting manner

(Figure 2).

Pathological state

Given these promising results in monitoring different levels of

consciousness in healthy subjects, TMS-EEG was subse-

quently performed in patients with DOC. In the two studies

published so far, a total of 30 patients were assessed

(15 UWS, 13 MCS and two patients with a locked-in

syndrome) during acute (less than 30 days post-brain injury),

sub-acute (between 1–3 months post-injury) and chronic

(more than 3 months post-anoxia and more than 1 year post-

trauma) periods [40, 56]. Nine of the 15 patients with an

unambiguous diagnosis of UWS showed a simple, local and

slow response to TMS, very similar to the one observed in

NREM sleep and general anaesthesia. The other six patients

with UWS did not show any response to TMS (one sub-acute

and five chronic patients). Interestingly, in the Rosanova et al.

[40] study, the only three patients who did not show any

response to TMS were the ones who suffered from anoxia.

Anoxic aetiology might be linked to some extent to the

absence of response in UWS patients. The exact implication

of the aetiology needs, however, to be further studied on a

larger population sample.

In MCS patients, 12 out of 13 showed complex responses

with changing patterns of cortical activation over time. These

responses were similar to the ones observed in the two

patients with locked-in syndrome (conscious patients who are

unable to produce any volitional motor output, except for eye

movements) and very similar to what is observed in normal

wakefulness. One patient in MCS, with an anoxic aetiology,

did not show any response to brain stimulation. Only one

brain area was stimulated in this patient, so one cannot

exclude that, if other brains areas had been perturbed, it would

have induced a complex and widespread response. Moreover,

the stimulus intensity was fixed as a percentage of the

maximum intensity deliverable and not based on actual

intensity at the cortical level; thus, the intensity necessary to

evoke a response may not have been reached in this patient.

Importantly, Ragazzoni et al. [56] compared the TMS-EEG

to other electrophysiological techniques such as somatosen-

sory evoked potentials, oddball auditory ERPs and EEG

power spectrum and showed that only TMS-EEG was able to

discriminate between patients in UWS and MCS at the

group level.

In acute settings, TMS-EEG was also able to detect, in

three patients with DOC, changes in cortical responses

matching those observed with behavioural evaluation [40].

The responses recorded when the patients were in an

unambiguous UWS were very similar to the one observed

previously in chronic UWS, as well as in NREM sleep and

general anaesthesia. On the other hand, once the patients

recovered signs of consciousness and, therefore, were

diagnosed as in a MCS, their responses to TMS looked like

those observed in patients with locked-in syndrome and in

conscious healthy subjects. Once the patients emerged from

the MCS (i.e. recovery of functional communication or

functional use of objects—EMCS), their brain responses to

TMS were also complex, as observed in other conscious states

(Figure 2). Interestingly, one patient who had improved to

MCS was behaviourally back in an UWS the day of the TMS

assessment, but complex and widespread brain responses

could still be detected, although, at the bedside, no sign of

consciousness could be observed. Two other acute UWS

patients who did not recover were also evaluated twice,

1 month apart. Each time, TMS triggered no response in one

patient and a simple, slow and local response in the other

patient. This suggests that, in the acute setting, TMS-EEG is

able to monitor recovery of consciousness or absence thereof

and is less sensitive to daily fluctuation in behavioural

diagnosis. However, it is still sensitive to the level of arousal,

as one patient in MCS showed simple and local slow wave in

response to TMS when sleeping, whereas complex brain

responses were recorded when awake [40]. In contrast,

transition from behavioural sleep to wakefulness was not

associated with a modification of the TMS responses in a

patient with an UWS (Figure 3).

Differences have also been observed between patients who

suffered mild traumatic brain injury and healthy subjects.

Some of the patients remained symptomatic, while some fully

recovered. None of them had, however, obvious structural

lesions on the MRI. In the symptomatic group, stimulation of

the dorsolateral prefrontal cortex elicited delayed peaks and

increased amplitude of some of the TMS-EEG components

[57]. In the asymptomatic group, stimulation of the same site

elicited responses of shortened latency. In both mild traumatic

brain injury groups, TMS recordings also revealed an

increased motor threshold (as defined by the minimum

TMS intensity required to elicit a 50 mV response on

electromyography in 5 out of 10 stimulations), as well

as differences in the effect of increasing stimulation intensity

as compared to healthy control. This increased motor

threshold has also been observed in patients in UWS and

MCS [58].

Patients with psychiatric disorders also show altered brain

responses to TMS. For instance, while stimulating frontal and

prefrontal brain regions, patients with schizophrenia demon-

strated a clear deficit in the production of fast EEG

oscillations, as compared to healthy subjects [59]. Frantseva

et al. [60] also found an abnormal widespread activation of

the EEG response to TMS in schizophrenia, which was

associated with higher global voltage between 400–750

milliseconds post-stimulation, as compared to healthy con-

trols [60]. Both findings revealed a correlation between EEG

gamma band abnormalities and positive symptoms in

schizophrenia.

TMS-EEG could also be an effective tool to assess

altered excitability and connectivity in neurodegenerative

diseases. Casarotto et al. [61] demonstrated a lower

frontal cortical excitability in patients with Alzheimer’s

disease compared to healthy elderly individuals. All these

1184 M. Napolitani et al. Brain Inj, 2014; 28(9): 1180–1189

applications of TMS-EEG clearly demonstrate its ability to

investigate the status of various cortical networks. However,

it is important to follow clear guidelines in order to

accurately perform TMS-EEG assessment, especially in

patients with DOC.

How to conduct a good experiment in patients withdisorders of consciousness?

As patients with DOC often have significant brain lesions,

one needs to carefully check the patient brain MRI or CT scan

prior to deciding the stimulation target. Stimulations should

be performed on intact brain areas, as direct stimulation of a

brain lesion is likely to trigger no or very little response,

hence increasing the risk of misdiagnosis. Stimulation of

several sites can avoid such a risk in the absence of recent

neuroimaging data. The experimenter has also to make sure

the patient stays awake for the duration of the TMS-EEG, for

example with the help of the CRS-R arousal protocol and by

positioning seated. If no responses or very little response is

observable online, intensity of the stimulation should be

increased, as responses are sometimes very subtle in this

population.

It is important to acknowledge the risk of inducing a

seizure in patients with a history of epilepsy. Although this is

rare and was mostly seen with high frequency stimulation

(more than 1 Hz) [62], patients with severe brain injuries

often had at least one episode of seizure and, thus, are at

higher risk to present another one. This issue should be

discussed with the local ethic committee, tempered by the

facts that (i) the stimulation parameters typically used are less

susceptible to trigger a seizure than high frequency repeated

stimulations, (ii) patients who had only one contextual seizure

and since then have been treated efficiently (i.e. have had

none since the treatment was started) are unlikely to present

another one and (iii) that TMS is performed while EEG is

recorded and, thus, a seizure would be rapidly noticed.

Moreover, TMS-EEG on patients with DOC is performed

most of the time in a setting where adequate management is

immediately available. The recorded response should also be

interpreted in light of the most recent theories of

consciousness.

Figure 3. (A) Effect of arousal on TMS-EEG findings in UWS and in MCS. Behavioural sleep or wakefulness does not influence the response observedin the patient in UWS, whereas in the patient in MCS there is a clear difference between eyes opened (EO) and eye closed (EC) conditions The responseunder the coil (black trace) and the stimulation site (white cross), with the significance threshold (p50.01, pink band) are depicted. The 10 mostactivated sources during the significant post-stimulus activation are plotted and colour-coded according to the location. (B) Effect of increasing level ofconsciousness. The same graphs are plotted for an acute UWS patient who improved behaviourally until emergence from MCS. Note thatthe complexity of the responses improves parallelly. (C) Typical findings in a patient with locked-in syndrom (LIS) showing very similar resultsto those observed in wakefulness, REM sleep and EMCS. EC, eyes closed; EO, eyes opened; UWS, Unresponsive wakefulness syndrome;MCS, Minimally conscious state; EMCS, emergence from minimally conscious state; LIS, Locked-in syndrome; TMS, transcranial magneticstimulation. Adapted from Rosanova et al. [40] with permission.

DOI: 10.3109/02699052.2014.920524 TMS-EEG in disorders of consciousness 1185

Theories of consciousness and the perturbationalapproach

In order to establish an objective and reliable marker of

human consciousness, there is a need to design a theoretical

framework that allows making hypothesis about the brain

mechanisms underlying conscious experiences. Although

different hypotheses exist, the most recent ones, such as the

global workspace theory of consciousness [63, 64], the

information integration theory of consciousness – IITC

[65]—and the cognitive binding theory [66], converge to

the necessity of a neural basis of consciousness matching the

temporal dynamic of conscious perception and cognitive

abilities (as described in Ortinski and Meador [67]), while

keeping in mind that these neural processes can be modulated

by other brain structures (e.g. the amygdala for emotional

stimuli [68]).

According to the IITC [65], consciousness can only arise if

its supporting neural network has the ability to differentiate

and integrate information. Indeed, the conscious events

experienced are at the same time different from any past

event and integrated into their context and are impossible to

isolate as such. At the neural level, this conscious event

results in a specific neuronal activation pattern among an

infinite number of possible states. For consciousness to

emerge, the brain, therefore, needs an integrated and

differentiated network, consisting of several specialized

modules that are effectively connected. To infer the brain

capacity for consciousness, one would then need to assess

effective connectivity between these brain regions and the

TMS-EEG technique seems to be the perfect tool to do so, as

seen above. The complex and widespread distributed response

observed in normal wakefulness, in patients in MCS or during

dreams, reflects that the underlying network is integrated

(the response spreads across distant cortical regions) and

specialized, providing information (the response is complex

and not stereotypical). On one hand, if the system’s modules

operate as independent sub-sets, being effectively discon-

nected from the rest of the network, no information can be

integrated; on the other hand, if there is no specialization and

the system behaves stereotypically, integration occurs without

differentiation. In both cases, the loss of information would

lead to a loss of consciousness. In NREM sleep, general

anaesthesia and UWS, the loss of integration gives rise to the

simple, local and short lasting response observed [35, 36, 40].

During general seizures, where the entire cortex is firing

anarchically, there is no specialization and, hence, no

information, leading to an alteration of consciousness [69].

The aforementioned results are mainly based on qualitative

comparisons between response shapes. To implement the

TMS-EEG technique in a clinical routine for the assessment

of consciousness, there is a need to objectively quantify those

responses. This has previously been done, to some extent,

with the SCD and SCS indices [42]. However, these indices

do not allow the direct comparison between individual

subjects and do not reflect complexity per se. A new

complexity measure, the Perturbational Complexity Index

(PCI), has been recently developed to assess the brain

capacity for consciousness. This ingenious measure is in

line with the IITC theory and allows comparison of different

subjects and different levels of consciousness [70]. Briefly,

based on TMS-EEG artefact-free recordings, statistically

significant sources are modelled and a binary matrix is

generated. This matrix is then compressed using algorithmic

complexity compression and normalized by the source

entropy, resulting in the PCI. PCI is low in cases of loss of

integration (as its time course would be quite short) and in

cases of loss of differentiation (because the information

contained would be extremely redundant and, thus, highly

compressible).

PCI is able to discriminate between conscious and

unconscious conditions regardless of the strength and extent

in duration of the cortical response, the stimulation site and

intensity (given that the latter is sufficient to evoke an EEG

response). PCI is also sensitive to graded changes of

consciousness (in different sleep stages, degree of general

anaesthesia and DOC). In conclusion, as shown in Figure 4,

this measure allows clear-cut differences in intracortical

effective connectivity between patients with severe brain

injury in UWS and those who recover consciousness

Figure 4. Pertubational Complexity Index (PCI) values in patients with DOC and healthy subjects. (A) PCI rise is matching increasing CRS-R andlevel of consciousness. UWS values are in the NREM sleep and general anaesthesia range, while EMCS and LIS are in the healthy awake subjectsrange. Patients in MCS show intermediate values, but never below the threshold of unconsciousness of healthy subjects (horizontal blue dashedline, PCI¼ 0.31). (B) Differences in group mean PCI values are significant between each pair of possible state of consciousness in DOC.CRS-R, Coma Recovery Scale-Revised; PCI, Perturbational Complexity Index; UWS, Unresponsive Wakefulness Syndrome; MCS, MinimallyConscious State; EMCS, Emergence from MCS; LIS, Locked-In Syndrome; NREM, Non-Rapid Eye Movement. *p¼ 0.002, **p¼ 0.0001. Adaptedfrom Casali et al. [70] with permission.

1186 M. Napolitani et al. Brain Inj, 2014; 28(9): 1180–1189

(MCS, EMCS, locked-in syndrome), representing for the first

time a reliable assessment of consciousness at the single

subject level [70].

Future perspectives

TMS-EEG and its derived measures have shown interesting

results in the field of DOC. PCI has demonstrated its consistent

ability in determining brain’s capacity for consciousness in a

significant number of subjects under many conditions.

However, it would be interesting to verify the consistency of

these results in a larger group of patients and also using

different set-ups. To be more easily implemented in a clinical

setting, the whole equipment should be more portable: easy to

bring and set up at the patients’ bedside. The pre-processing

and analyses of TMS-EEG derived measures should ideally be

more straightforward, possibly automated and a standard

procedure should be available. This would allow different

centres to obtain directly comparable measures of conscious-

ness and to elaborate guidelines for the diagnosis of DOC using

more than behavioural scales. Larger longitudinal studies of

patients with DOC are also needed to confirm the sensitivity to

graded changes in consciousness, as only three patients have

been studied during the course of their recovery, so far.

Early prognostic markers should be implemented to identify

earlier patients who are more likely to recover consciousness.

The effect of brain injury aetiology on TMS-EEG response and

its implication on the prognosis could also be studied.

Similarly, TMS-EEG should be used to assess the effect of

pharmacological treatments including zolpidem, amantadine

and apomorphine, which have been shown to promote

functional recovery in a sub-set of patients with DOC

[71–73]. This would allow researchers to better understand

the mechanisms by which such treatments work and perhaps

determine early on the patients who are likely to respond.

Repetitive TMS, which has not been discussed in this review,

can modulate brain excitability and potentially connectivity.

These effects could also be quickly assessed by TMS-EEG.

Finally, TMS-EEG results could be combined with those

obtained with other techniques in order to study neural

correlates of consciousness. For example, correlating effective

connectivity with structural connectivity (using diffusion

tensor imaging) or functional connectivity (using EEG or

fMRI [74]). In fact, structural data would provide prior

information as input for the source-modelling algorithm,

possibly increasing its power and making it closer to reality,

as effective connectivity is supported by structural connectiv-

ity [34]. Intracranial EEG recordings in patients evaluated for

pharmacoresistant epilepsy, for instance, would also extend

the spatial capacity of this approach and allow studying of

the deeper structures of the thalamocortical networks.

Previous studies have demonstrated the safety of using TMS

with implanted electrodes, although some recommendations

have been released; such as avoiding stimulation near

electrodes loops and avoiding repetitive TMS in a chronic

setting due to lack of data on potential tissue damage [75].

Conclusion

The concomitant development of combined TMS-EEG

technique and new theories of consciousness has led to a

promising evolution in the way of investigating brain

processes. In acute and chronic DOC, TMS-EEG is uncover-

ing new aspects of the brain physiology. Most importantly, for

the first time, one has the potential to assess the residual level

of consciousness of patients with severe brain injuries at the

single subject level and at the bedside. Further developments

such as the PCI are likely to improve the feasibility of a daily

practical application of TMS in evaluation of consciousness.

However, as with every new technique, several questions

remain to be answered, such as the exact sensitivity and

specificity of this tool or its potential prognostic value and the

exact physiology of TMS responses, which has not yet been

determined.

Acknowledgements

The authors would like to thank Simone Sarasso and

Chanyoung Kang for their constructive comments about this

manuscript.

Declaration of interest

This research was funded by the University and University

Hospital of Liege, the Leon Fredericq Funds, the Belgian

National Funds for Scientific Research (FRS-FNRS), the

European Commission (COST, DISCOS, MINDBRIDGE,

DECODER), the James McDonnell Foundation, the Mind

Science Foundation, Wallonie-Bruxelles International (WBI)

the French Speaking Community Concerted Research Action

(ARC 06/11-340), the Belgian American Educational

Foundation (BAEF) the Foundation Medicale Reine

Elisabeth, the Public Utility Foundation ‘Universite

Europeenne du Travail’ and ‘Fondazione Europea di

Ricerca Biomedica’. OG received support from NIH grant

MH064498 and MH095984 to Bradley R. Postle and Giulio

Tononi. OB is a research fellow, OG a postdoctoral researcher

and SL a research director at the FNRS.

References

1. Laureys S, Celesia GG, Cohadon F, Lavrijsen J, Leon-Carrion J,Sannita WG, Sazbon L, Schmutzhard E, von Wild KR, Zeman A,et al. Unresponsive wakefulness syndrome: A new name for thevegetative state or apallic syndrome. BMC Medicine 2010;8:68.

2. Giacino JT, Ashwal S, Childs N, Cranford R, Jennett B, Katz DI,Kelly JP, Rosenberg JH, Whyte J, Zafonte RD, et al. The minimallyconscious state: Definition and diagnostic criteria. Neurology 2002;58:349–353.

3. Schnakers C, Vanhaudenhuyse A, Giacino J, Ventura M, Boly M,Majerus S, Moonen G, Laureys S. Diagnostic accuracy of thevegetative and minimally conscious state: Clinical consensus versusstandardized neurobehavioral assessment. BMC Neurology 2009;9:35.

4. Giacino JT, Kalmar K, Whyte J. The JFK Coma Recovery Scale-Revised: Measurement characteristics and diagnostic utility.Archives of Physical Medicine and Rehabilitation 2004;85:2020–2029.

5. Massimini M, Boly M, Casali A, Rosanova M, Tononi G.A perturbational approach for evaluating the brain’s capacity forconsciousness. Progress in Brain Research 2009;177:201–214.

6. Virtanen J, Ruohonen J, Naatanen R, Ilmoniemi RJ.Instrumentation for the measurement of electric brain responsesto transcranial magnetic stimulation. Medical & BiologicalEngineering & Computing 1999;37:322–326.

7. Majerus S, Bruno MA, Schnakers C, Giacino JT, Laureys S. Theproblem of aphasia in the assessment of consciousness in brain-damaged patients. Progress in Brain Research 2009;177:49–61.

DOI: 10.3109/02699052.2014.920524 TMS-EEG in disorders of consciousness 1187

8. Boly M, Coleman MR, Davis MH, Hampshire A, Bor D,Moonen G, Maquet PA, Pickard JD, Laureys S, Owen AM. Whenthoughts become action: An fMRI paradigm to study volitionalbrain activity in non-communicative brain injured patients.NeuroImage 2007;36:979–992.

9. Owen AM, Coleman MR, Boly M, Davis MH, Laureys S,Pickard JD. Using functional magnetic resonance imaging todetect covert awareness in the vegetative state. Archives ofNeurology 2007;64:1098–1102.

10. Monti MM, Vanhaudenhuyse A, Coleman MR, Boly M,Pickard JD, Tshibanda L, Owen AM, Laureys S. Willful modula-tion of brain activity in disorders of consciousness. TheNew England Journal of Medicine 2010;362:579–589.

11. Bardin JC, Fins JJ, Katz DI, Hersh J, Heier LA, Tabelow K,Dyke JP, Ballon DJ, Schiff ND, Voss HU. Dissociations betweenbehavioural and functional magnetic resonance imaging-basedevaluations of cognitive function after brain injury. Brain 2011;134:769–782.

12. Vogel D, Markl A, Yu T, Kotchoubey B, Lang S, Muller F.Can mental imagery functional magnetic resonance imaging predictrecovery in patients with disorders of consciousness? Archives ofPhysical Medicine and Rehabilitation 2013;10:1891–1898.

13. Boly M, Faymonville ME, Peigneux P, Lambermont B, Damas P,Del Fiore G, Degueldre C, Franck G, Luxen A, Lamy M, et al.Auditory processing in severely brain injured patients: Differencesbetween the minimally conscious state and the persistent vegetativestate. Archives of Neurology 2004;61:233–238.

14. Schiff ND, Ribary U, Moreno DR, Beattie B, Kronberg E,Blasberg R, Giacino J, McCagg C, Fins JJ, Llinas R, et al.Residual cerebral activity and behavioural fragments can remain inthe persistently vegetative brain. Brain 2002;125:1210–1234.

15. Laureys S, Faymonville ME, Peigneux P, Damas P, Lambermont B,Del Fiore G, Degueldre C, Aerts J, Luxen A, Franck G, et al.Cortical processing of noxious somatosensory stimuli in thepersistent vegetative state. NeuroImage 2002;17:732–741.

16. Vanhaudenhuyse A, Noirhomme Q, Tshibanda LJ, Bruno MA,Boveroux P, Schnakers C, Soddu A, Perlbarg V, Ledoux D,Brichant JF, et al. Default network connectivity reflects the level ofconsciousness in non-communicative brain-damaged patients.Brain 2010;133:161–171.

17. Demertzi A, Gomez F, Crone JS, Vanhaudenhuyse A, Tshibanda L,Noirhomme Q, Thonnard M, Charland-Verville V, Kirsch M,Laureys S, et al. Multiple fMRI system-level baseline connectivityis disrupted in patients with consciousness alterations. Cortex;A Journal Devoted to the Study of the Nervous System andBehavior 2014;52:35–46.

18. Jox RJ, Bernat JL, Laureys S, Racine E. Disorders of conscious-ness: Responding to requests for novel diagnostic and therapeuticinterventions. Lancet Neurology 2012;11:732–738.

19. Dehaene S, Changeux JP. Experimental and theoretical approachesto conscious processing. Neuron 2011;70:200–227.

20. Cruse D, Chennu S, Chatelle C, Bekinschtein TA,Fernandez-Espejo D, Pickard JD, Laureys S, Owen AM. Bedsidedetection of awareness in the vegetative state: A cohort study.Lancet 2011;378:2088–2094.

21. Schnakers C, Perrin F, Schabus M, Hustinx R, Majerus S,Moonen G, Boly M, Vanhaudenhuyse A, Bruno MA, Laureys S.Detecting consciousness in a total locked-in syndrome: An activeevent-related paradigm. Neurocase 2009;15:271–277.

22. Naci L, Monti MM, Cruse D, Kubler A, Sorger B, Goebel R,Kotchoubey B, Owen AM. Brain-computer interfaces for commu-nication with nonresponsive patients. Annals of Neurology 2012;72:312–323.

23. Myles PS, Leslie K, McNeil J, Forbes A, Chan MT. Bispectralindex monitoring to prevent awareness during anaesthesia: TheB-Aware randomised controlled trial. Lancet 2004;363:1757–1763.

24. Schnakers C, Ledoux D, Majerus S, Damas P, Damas F,Lambermont B, Lamy M, Boly M, Vanhaudenhuyse A,Moonen G, et al. Diagnostic and prognostic use of bispectralindex in coma, vegetative state and related disorders. Brain Injury2008;22:926–931.

25. Landsness E, Bruno MA, Noirhomme Q, Riedner B, Gosseries O,Schnakers C, Massimini M, Laureys S, Tononi G, Boly M.Electrophysiological correlates of behavioural changes in vigilance

in vegetative state and minimally conscious state. Brain 2011;134:2222–2232.

26. Gosseries O, Schnakers C, Ledoux D, Vanhaudenhuyse A,Bruno MA, Demertzi A, Noirhomme Q, Lehembre R, Damas P,Goldman S, et al. Automated EEG entropy measurements in coma,vegetative state/unresponsive wakefulness syndrome and minimallyconscious state. Functional Neurology 2011;26:25–30.

27. Tononi G, Sporns O, Edelman GM. A measure for braincomplexity: Relating functional segregation and integration in thenervous system. Proceedings of the National Academy of Sciencesof the United States of America 1994;91:5033–5037.

28. Seth AK. Causal connectivity of evolved neural networks duringbehavior. Network 2005;16:35–54.

29. Granger CW. Investigating causal relations by Econometric Modelsand Cross-spectral Methods. Econometrica 1969;37:424–238.

30. Balduzzi D. Estimating integrated information with TMS pulsesduring wakefulness, sleep, and under anesthesia. Conferenceproceedings: Annual International Conference of the IEEEEngineering in Medicine and Biology Society. IEEEEngineering in Medicine and Biology Society. Conference2011:4717–4720.

31. Roth BJ, Saypol JM, Hallett M, Cohen LG. A theoreticalcalculation of the electric field induced in the cortex duringmagnetic stimulation. Electroencephalography and ClinicalNeurophysiology 1991;81:47–56.

32. Ruohonen J, Ilmoniemi RJ. Modeling of the stimulating fieldgeneration in TMS. Electroencephalography and ClinicalNeurophysiology Supplement 1999;51:30–40.

33. Maccabee PJ, Amassian VE, Eberle LP, Cracco RQ. Magnetic coilstimulation of straight and bent amphibian and mammalianperipheral nerve in vitro: Locus of excitation. The Journal ofPhysiology 1993;460:201–219.

34. Friston KJ. Functional and effective connectivity: A review.Brain Connectivity 2011;1:13–36.

35. Massimini M, Ferrarelli F, Huber R, Esser SK, Singh H, Tononi G.Breakdown of cortical effective connectivity during sleep. Science2005;309:2228–2232.

36. Ferrarelli F, Massimini M, Sarasso S, Casali A, Riedner BA,Angelini G, Tononi G, Pearce RA. Breakdown in cortical effectiveconnectivity during midazolam-induced loss of consciousness.Proceedings of the National Academy of Sciences of the UnitedStates of America 2010;107:2681–2686.

37. Bonato C, Miniussi C, Rossini PM. Transcranial magnetic stimu-lation and cortical evoked potentials: A TMS/EEG co-registrationstudy. Clinical Neurophysiology: Official Journal of theInternational Federation of Clinical Neurophysiology 2006;117:1699–1707.

38. Esser SK, Huber R, Massimini M, Peterson MJ, Ferrarelli F,Tononi G. A direct demonstration of cortical LTP in humans:A combined TMS/EEG study. Brain Research Bulletin 2006;69:86–94.

39. Mutanen T, Maki H, Ilmoniemi RJ. The effect of stimulusparameters on TMS-EEG muscle artifacts. Brain Stimulation2013;6:371–376.

40. Rosanova M, Gosseries O, Casarotto S, Boly M, Casali AG,Bruno MA, Mariotti M, Boveroux P, Tononi G, Laureys S, et al.Recovery of cortical effective connectivity and recovery ofconsciousness in vegetative patients. Brain 2012;135:1308–1320.

41. Casarotto S, Romero Lauro LJ, Bellina V, Casali AG, Rosanova M,Pigorini A, Defendi S, Mariotti M, Massimini M. EEG responses toTMS are sensitive to changes in the perturbation parameters andrepeatable over time. PLoS One 2010;5:e10281.

42. Casali AG, Casarotto S, Rosanova M, Mariotti M, Massimini M.General indices to characterize the electrical response of thecerebral cortex to TMS. NeuroImage 2010;49:1459–1468.

43. Johnson JS, Kundu B, Casali AG, Postle BR. Task-dependentchanges in cortical excitability and effective connectivity:A combined TMS-EEG study. Journal of Neurophysiology 2012;107:2383–2392.

44. Kundu B, Sutterer DW, Emrich SM, Postle BR. Strengthenedeffective connectivity underlies transfer of working memorytraining to tests of short-term memory and attention. The Journalof Neuroscience: The Official Journal of the Society forNeuroscience 2013;33:8705–8715.

1188 M. Napolitani et al. Brain Inj, 2014; 28(9): 1180–1189

45. Akaishi R, Morishima Y, Rajeswaren VP, Aoki S, Sakai K.Stimulation of the frontal eye field reveals persistent effectiveconnectivity after controlled behavior. The Journal ofNeuroscience: The Official Journal of the Society forNeuroscience 2010;30:4295–4305.

46. Morishima Y, Akaishi R, Yamada Y, Okuda J, Toma K, Sakai K.Task-specific signal transmission from prefrontal cortex in visualselective attention. Nature Neuroscience 2009;12:85–91.

47. Nikulin VV, Kicic D, Kahkonen S, Ilmoniemi RJ. Modulation ofelectroencephalographic responses to transcranial magnetic stimu-lation: Evidence for changes in cortical excitability related tomovement. The European Journal of Neuroscience 2003;18:1206–1212.

48. Massimini M, Ferrarelli F, Murphy M, Huber R, Riedner B,Casarotto S, Tononi G. Cortical reactivity and effective connect-ivity during REM sleep in humans. Cognitive Neuroscience 2010;1:176–183.

49. Rosanova M, Casali A, Bellina V, Resta F, Mariotti M,Massimini M. Natural frequencies of human corticothalamiccircuits. The Journal of Neuroscience: The official Journal of theSociety for Neuroscience 2009;29:7679–7685.

50. Huber R, Maki H, Rosanova M, Casarotto S, Canali P, Casali AG,Tononi G, Massimini M. Human cortical excitability increases withtime awake. Cerebral Cortex 2013;23:332–338.

51. Mattavelli G, Rosanova M, Casali AG, Papagno C,Romero Lauro LJ. Top-down interference and cortical responsive-ness in face processing: A TMS-EEG study. NeuroImage 2013;76:24–32.

52. Kahkonen S, Wilenius J, Nikulin VV, Ollikainen M, Ilmoniemi RJ.Alcohol reduces prefrontal cortical excitability in humans:A combined TMS and EEG study. Neuropsychopharmacology2003;28:747–754.

53. Hobson JA, Pace-Schott EF, Stickgold R. Dreaming and the brain:Toward a cognitive neuroscience of conscious states. Behav BrainSci 2000;23:793–842.

54. Massimini M, Ferrarelli F, Esser SK, Riedner BA, Huber R,Murphy M, Peterson MJ, Tononi G. Triggering sleep slow waves bytranscranial magnetic stimulation. Proceedings of the NationalAcademy of Sciences of the United States of America 2007;104:8496–8501.

55. Tanelian DL, Kosek P, Mody I, MacIver MB. The role of theGABAA receptor/chloride channel complex in anesthesia.Anesthesiology 1993;78:757–776.

56. Ragazzoni A, Pirulli C, Veniero D, Feurra M, Cincotta M,Giovannelli F, Chiaramonti R, Lino M, Rossi S, Miniussi C.Vegetative versus Minimally Conscious States: A study usingTMS-EEG, sensory and event-related potentials. PLoS One 2013;8:e57069.

57. Tallus J, Lioumis P, Hamalainen H, Kahkonen S, Tenovuo O.TMS-EEG responses in recovered and symptomatic mild traumaticbrain injury. Journal of Neurotrauma 2013;30:1270–1277.

58. Lapitskaya N, Gosseries O, De Pasqua V, Pedersen AR, Nielsen JF,de Noordhout AM, Laureys S. Abnormal corticospinal excitabilityin patients with disorders of consciousness. Brain Stimulation 2013;6:590–597.

59. Ferrarelli F, Massimini M, Peterson MJ, Riedner BA, Lazar M,Murphy MJ, Huber R, Rosanova M, Alexander AL, Kalin N, et al.Reduced evoked gamma oscillations in the frontal cortex inschizophrenia patients: A TMS/EEG study. The American Journalof Psychiatry 2008;165:996–1005.

60. Frantseva M, Cui J, Farzan F, Chinta LV, Perez Velazquez JL,Daskalakis ZJ. Disrupted cortical conductivity in schizophrenia:TMS-EEG Study. Cerebral Cortex 2014;24:211–221.

61. Casarotto S, Maatta S, Herukka SK, Pigorini A, Napolitani M,Gosseries O, Niskanen E, Kononen M, Mervaala E, Rosanova M,et al. Transcranial magnetic stimulation-evoked EEG/corticalpotentials in physiological and pathological aging. Neuroreport2011;22:592–597.

62. Wassermann EM. Risk and safety of repetitive transcranialmagnetic stimulation: Report and suggested guidelines from theInternational Workshop on the Safety of Repetitive TranscranialMagnetic Stimulation, June 5–7, 1996. Electroencephalography andClinical Neurophysiology 1998;108:1–16.

63. Baars BJ. The conscious access hypothesis: Origins and recentevidence. Trends in Cognitive Sciences 2002;6:47–52.

64. Dehaene S, Naccache L. Towards a cognitive neuroscience ofconsciousness: Basic evidence and a workspace framework.Cognition 2001;79:1–37.

65. Tononi G. An information integration theory of consciousness.BMC Neuroscience 2004;5:42.

66. Mashour GA. Consciousness unbound: Toward a paradigm ofgeneral anesthesia. Anesthesiology 2004;100:428–433.

67. Ortinski P, Meador KJ. Neuronal mechanisms of consciousawareness. Archives of Neurology 2004;61:1017–1020.

68. Amting JM, Greening SG, Mitchell DG. Multiple mechanisms ofconsciousness: The neural correlates of emotional awareness. TheJournal of Neuroscience: The Official Journal of the Society forNeuroscience 2010;30:10039–10047.

69. McPherson A, Rojas L, Bauerschmidt A, Ezeani CC, Yang L,Motelow JE, Farooque P, Detyniecki K, Giacino JT,Blumenfeld H. Testing for minimal consciousness in complexpartial and generalized tonic-clonic seizures. Epilepsia 2012;53:e180–e183.

70. Casali AJ, Gosseries O, Rosanova M, Boly M, Sarasso S, CasaliKR, Casarotto S, Bruno M-A, Laureys S, Tononi G, Massimini M.A theoretically based index of consciousness independent ofsensory processing and behavior. Science Translational Medicine2013;5:198ra105. doi: 10.1126/scitranslmed.3006294.

71. Giacino JT, Whyte J, Bagiella E, Kalmar K, Childs N, Khademi A,Eifert B, Long D, Katz DI, Cho S, et al. Placebo-controlled trial ofamantadine for severe traumatic brain injury. The New EnglandJournal of Medicine 2012;366:819–826.

72. Clauss R, Nel W. Drug induced arousal from the permanentvegetative state. NeuroRehabilitation 2006;21:23–28.

73. Fridman EA, Krimchansky BZ, Bonetto M, Galperin T, Gamzu ER,Leiguarda RC, Zafonte R. Continuous subcutaneous apomorphinefor severe disorders of consciousness after traumatic brain injury.Brain Injury 2010;24:636–641.

74. Peters JC, Reithler J, Schuhmann T, de Graaf T, Uludag K,Goebel R, Sack AT. On the feasibility of concurrent humanTMS-EEG-fMRI measurements. Journal of Neurophysiology 2013;109:1214–1227.

75. Shimojima Y, Morita H, Nishikawa N, Kodaira M, Hashimoto T,Ikeda S. The safety of transcranial magnetic stimulation with deepbrain stimulation instruments. Parkinsonism & Related Disorders2010;16:127–131.

76. Sarasso S, Rosanova M, Casali AG, Casarotto S, Fecchio M, BolyM, Gosseries O, Tononi G, Laureys S, Massimini M. Quantifyingcortical EEG responses to TMS in (un)consciousness.Clinical EEG and Neuroscience 2014;45:40–49. doi: 10.1177/1550059413513723.

DOI: 10.3109/02699052.2014.920524 TMS-EEG in disorders of consciousness 1189