This article was downloaded by: [UQ Library]

On: 04 November 2014, At: 23:29

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Cognitive Neuroscience
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/pcns20

Disorders of consciousness: Moving from passive to

resting state and active paradigms
a a a a b a a
M. A. Bruno , A. Soddu , A. Demertzi , S. Laureys , O. Gosseries , C. Schnakers ,
a a a a a
M. Boly , Q. Noirhomme , M. Thonnard , C. Chatelle & A. Vanhaudenhuyse
a
University and University Hospital of Liège , Liège, Belgium
b
University Hospital of Liège , Liège, Belgium
Published online: 02 Jul 2010.

To cite this article: M. A. Bruno , A. Soddu , A. Demertzi , S. Laureys , O. Gosseries , C. Schnakers , M. Boly , Q.
Noirhomme , M. Thonnard , C. Chatelle & A. Vanhaudenhuyse (2010) Disorders of consciousness: Moving from passive to
resting state and active paradigms, Cognitive Neuroscience, 1:3, 193-203, DOI: 10.1080/17588928.2010.485677

To link to this article: http://dx.doi.org/10.1080/17588928.2010.485677

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 COGNITIVE NEUROSCIENCE, 2010, 1 (3), 193–203

Disorders of consciousness: Moving from passive

PCNS

to resting state and active paradigms

M. A. Bruno1, A. Soddu1, A. Demertzi1, S. Laureys1,2, O. Gosseries1, C. Schnakers1,

Disorders of Consciousness

M. Boly1, Q. Noirhomme1, M. Thonnard1, C. Chatelle1, and A. Vanhaudenhuyse1

1
University and University Hospital of Liège, Liège, Belgium
2
University Hospital of Liège, Liège, Belgium
Downloaded by [UQ Library] at 23:29 04 November 2014

Following coma, some patients will recover wakefulness without signs of consciousness (i.e., vegetative state) or
may show nonreflexive movements but with no ability for functional communication (i.e., minimally conscious
state). Currently, there remains a high rate of misdiagnosis of the vegetative state. The increasing use of fMRI and
EEG tools permits the clinical characterization of these patients to be improved. We first discuss “resting metabolism”
and “passive activation” paradigms, used in neuroimaging and evoked potential studies, which merely identify
neural activation reflecting “automatic” processing—that is, occurring without the patient’s willful intervention.
Secondly, we present an alternative approach consisting of instructing subjects to imagine well-defined sensory-
motor or cognitive-mental actions. This strategy reflects volitional neural activation and, hence, witnesses awareness.
Finally, we present results on blood-oxgen-level-dependent “default mode network”/resting state studies that
might be a promising tool in the diagnosis of these challenging patients.

Keywords: Consciousness; Vegetative state; Resting state; fMRI; Active paradigms.

INTRODUCTION regain consciousness in the first month after brain

injury. However, if patients show no signs of awareness
Following severe brain damage, patients classically 1 year after a traumatic brain injury or 3 months after
evolve through different clinical stages before recov- anoxic accident, the chances of recovering functional
ering consciousness. Coma is usually a transient con- communication are close to zero and the patient is
dition that will last no longer than a few days or considered in a permanent VS (Multi-Society Task
weeks and is defined as an “unarousable unrespon- Force on PVS, 1994). Those who recover from VS
siveness” (Posner, Saper, Schiff, & Plum, 2007). typically progress to a minimally conscious state
After some days or weeks, comatose patients may (MCS), which is characterized by minimal but definite
open their eyes. When this return to “wakefulness” is behavioral evidence of awareness of self and/or the
accompanied by reflexive motor activity only, the environment (Giacino et al., 2002). Like the VS, the
condition is called vegetative state (VS; Multi-Society MCS may be chronic and sometimes permanent.
Task Force on PVS, 1994). Many patients in VS However, at present, no time-windows for “permanent

Correspondence should be addressed to: S. Laureys, Coma Science Group, Cyclotron Research Center, University and University Hospital
of Liège, Sart-Tilman B30, Liège, Belgium. E-mail: [email protected]
M. A. Bruno and A. Soddu contributed equally to the manuscript.
SL is Senior Research Associate; AS, MB and QN are Post-doctoral Fellows; and MAB is Research Fellow at the Fonds de la
Recherche Scientifique (FRS). This research was supported by the Fonds de la Recherche Scientifique (FRS), the European Commission
(DISCOS, Mindbridge, DECODER, & CATIA), Concerted Research Action (ARC-06/11-340), The McDonnell Foundation, and The
Mind Science Foundation.

© 2010 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business
www.psypress.com/cognitiveneuroscience DOI: 10.1080/17588928.2010.485677
 194 BRUNO ET AL.

MCS” have been agreed upon. Reliable and consistent emission tomography (PET), functional magnetic
interactive communication and/or functional use of resonance imaging (fMRI), and electrophysiological
objects indicate the next boundary in the course of techniques for the assessment of patients with disor-
recovery: emergence from MCS (Giacino et al., ders of consciousness, and present new data on rest-
2002). Finally, some patients may awake from their ing state brain activity in VS patients.
coma but be unable to move or speak: the only way
they have to express their consciousness is by small
eye movements. This state is called locked-in syn- GLOBAL RESTING METABOLISM
drome (LIS) and is characterized by quadriplegia and
anarthria with general preservation of cognition Using PET, Levy et al. (1987) first showed that
(Schnakers et al., 2008a) and an elementary mode patients in a VS suffer from a massive cerebral meta-
of communication that uses vertical or lateral eye bolic reduction, estimated to be 40–50% of normal
movements (American Congress of Rehabilitation values. These results have been repeatedly confirmed
Medicine, 1995). by others (Laureys et al., 1999a; Rudolf, Ghaemi,
Behavioral assessment is one of the main meth- Haupt, Szelies, & Heiss, 1999; Tommasino, Grana,
ods used to detect awareness in severely brain Lucignani, Torri, & Fazio, 1995) in VS of different
injured patients recovering from coma. Clinical etiologies and duration. In patients with an LIS, overall
Downloaded by [UQ Library] at 23:29 04 November 2014

practice shows that recognizing unambiguous signs supratentorial cerebral metabolism has been shown to
of conscious perception in such patients can be very be preserved partially or fully (Levy et al., 1987),
challenging. Indeed, behavioral assessment is com- whereas in comatose patients a decrease of 45% in
plicated by the presence of motor/language impair- cerebral metabolism has been observed (Laureys
ment, tracheotomy, fluctuating arousal level or et al., 2001). However, a global depression of cerebral
ambiguous and rapidly habituating responses, which metabolism is not specific to VS or coma only. In
may partly explain the high frequency of misdiagno- slow wave sleep, overall brain metabolism also
sis (up to 43%) in these patients (Schnakers et al., decreases approximately to 40% of normal waking
2009b). This underestimation of the patient’s level values while in REM-sleep metabolism returns to
of consciousness can also be explained by other normal values (Maquet et al., 1990). Another example
factors such as poor expertise in behavioral assess- of transient metabolic depression is observed during
ment or the use of insensitive behavioral assessment general anesthesia, which is characterized by compa-
tools (e.g., using a mirror when evaluating visual rable reduction in cortical metabolism to that
pursuit, behavior that is considered as one of the first observed in VS (Alkire et al., 1999). However, we
clinical signs differentiating MCS from VS; have shown that the relationship between global levels
Vanhaudenhuyse, Schnakers, Bredart, and Laureys, of brain function and the presence or absence of
2008b). LIS may also be misdiagnosed as coma or awareness is not absolute; rather, some areas in the
VS possibly due to the rarity of this syndrome, the brain seem more important than others for the emer-
difficulty of recognizing unambiguous signs of con- gence of awareness (Laureys, Faymonville, Moonen,
sciousness, the fluctuation of vigilance in the acute Luxen, & Maquet, 2000c; Laureys et al., 1999a).
setting or by additional cognitive or sensory deficits, Indeed, VS patients who subsequently recovered con-
such as deafness (Bruno et al., 2009; Schnakers et sciousness did not show substantial changes in global
al., 2008a; Smart et al., 2008). Misdiagnosis can lead metabolic rates for glucose metabolism (Laureys,
to grave consequences, especially in end-of-life Lemaire, Maquet, Phillips, & Franck, 1999b), and
decision-making. Contrary to patients in VS, those some awake healthy volunteers have global brain
in MCS retain some capacity for cognitive process- metabolism values comparable to those observed in
ing (see next section). Moreover, the prognosis of some patients in a VS (Laureys, 2005). In VS, a dys-
MCS patients is significantly more favorable relative function was found not in the whole brain but in a
to those in VS (Giacino, 2004a). Clinical attitudes wide frontoparietal network encompassing polymodal
and decisions about end-of-life and pain, therefore, associative cortices: bilateral lateral frontal regions;
are likely to be influenced by whether one is parieto-temporal and posterior parietal areas; mesiof-
diagnosed as a VS or a MCS patient (Demertzi et al., rontal, posterior cingulated, and precuneal cortices
2009). New technical methods offer specific assess- (Laureys et al., 1999a, 2004a). Posterior cingulate and
ment procedures and the possibility to determine adjacent precuneal cortices were reported to differen-
objectively whether an unresponsive patient is tiate MCS from VS patients (Laureys, Owen, &
aware without explicit verbal or motor response. We Schiff, 2004a; intermediate metabolism in MCS,
here review neuroimaging studies using positron higher than in VS but lower than in conscious
 DISORDERS OF CONSCIOUSNESS 195

controls). More recently, a study on patients in the with emotional valence (the voice of the patient’s
chronic stage of traumatic diffuse brain injury mother compared with an unfamiliar voice) were
showed a bilateral hypometabolism in the medial also shown to activate amygdala in a traumatic
prefrontal regions, the medial frontobasal regions, MCS patient (Bekinschtein et al., 2004). Similarly,
the cingulate gyrus, and the thalamus. At the group Schiff et al. (2005), in two MCS patients, showed
level, these regions were more hypometabolic in VS selective activation in components of the cortical
than MCS patients, but MCS still showed less acti- language networks during presentation of narratives
vation than patients who have emerged from this read by a familiar voice and containing personally
state and who recovered the ability to communicate meaningful content. Such context-dependent
(Nakayama, Okumura, Shinoda, Nakashima, & higher-order auditory processing shows that con-
Iwama, 2006). Specifically, awareness seems tent does matter when talking to MCS patients.
related not exclusively to the activity in the fron- Finally, some exceptional VS patients may also
toparietal network but, as importantly, to the func- show higher atypical level of cortical activation in
tional connectivity within this network and the response to auditory stimulations, and this was pro-
thalami. Indeed, long-range cortico-cortical and posed to be a surrogate marker of good prognosis
cortico-thalamo-cortical “functional disconnec- (Di et al., 2007).
tions” could be identified in the VS (Laureys et al.,
Downloaded by [UQ Library] at 23:29 04 November 2014

1999a, 2000b).
Event-related potentials

PASSIVE PARADIGMS: CEREBRAL Event-related potentials (ERPs) have been used for a
ACTIVATION FROM EXTERNAL long time to assess comatose patients. Early compo-
STIMULATION nents of these potentials arising within 100 ms are
known to persist even in unconscious states. The
Functional neuroimaging later components of exogenous potentials and other
so-called “endogenous” ERP components (e.g.,
Somatosensory, auditory, and visual perceptions P300) are more reliably related to the (unconscious
are conscious experiences; thus, the wakeful or conscious) cognitive processing of the informa-
unconsciousness of VS patients, by definition, pre- tion, and less frequently observed in disorders of
cludes these experiences. However, the absence of consciousness (Vanhaudenhuyse, Laureys, & Perrin,
a behavioral response cannot be taken as an abso- 2008a). In several studies, ERPs were used to evalu-
lute proof of the absence of consciousness. Several ate the integrity of detection of noncommunicative
fMRI activation studies in VS (Coleman et al., patients’ own name, in order to assess the possible
2007; Di et al., 2007; Fernandez-Espejo et al., preservation of residual linguistic and self-processing
2008; Moritz et al., 2001; Staffen, Kronbichler, in these patients. Differential P300 wave to the own
Aichhorn, Mair, and Ladurner, 2006) have con- name (as compared to other names) was observed in
firmed previous PET studies showing preserved LIS patients (Perrin et al., 2006), which is not
activation of “lower level” primary sensory cortices surprising since their cognitive functions and their
which are disconnected from “higher order” associ- linguistic comprehension remain preserved (Onofrj,
ative cortical networks employing auditory (Boly et Thomas, Paci, Scesi, and Tombari, 1997; Schnakers
al., 2004; Laureys et al., 2000a; Owen et al., 2002), et al., 2008a). In MCS, results suggested that the
somatosensory (Boly et al., 2008a), or visual auditory system was relatively preserved in response
(Menon et al., 1998; Owen et al., 2002) stimula- to passive tones and language stimulation, implying
tions. Similar studies in PET reported that MCS that these patients are able to detect salient words
patients showed a more widespread activation than (Perrin et al., 2006). Most surprisingly, some VS
VS, with a cortico-cortical functional connectivity patients emitted a differential P300, although
more efficient in MCS compared to VS (Boly et al., delayed as compared to age-matched controls (Perrin
2004). These results confirmed that MCS patients et al., 2006). These last results showed that the P300
may experience pain and hence should systemati- wave resulting from a passive paradigm is not useful
cally receive appropriate analgesic treatment. to successfully distinguish unconscious from
Moreover, stimuli with emotional valence (baby conscious patients. However, in most studies, the
cries and own name) were shown to induce a much presence of a P300 correlated with favorable out-
more widespread activation than did meaningless come in comatose patients (Vanhaudenhuyse et al.,
noise in the MCS (Laureys et al., 2004b). Stimuli 2008a).
 196 BRUNO ET AL.

ACTIVE PARADIGMS: COMMAND patients in VS may be conscious. Interestingly, when

FOLLOWING IN NONCOMMUNICATIVE re-examined six months later, the patient showed
PATIENTS inconsistent signs of consciousness. The most likely
explanation of these results is that the patient was
Functional neuroimaging already beginning the transition to the MCS at the
time of the experiment. This study also highlights the
In the absence of a full understanding of the neural importance of fMRI as a potentially good marker for
correlates of consciousness, even a near-normal acti- both diagnosis and prognosis (Di et al., 2007).
vation in response to passive stimulations cannot be
considered a proof of the presence of awareness.
Instead, all that can be inferred is that a specific brain ERPs
region is, to some degree, still able to perceive and
process relevant sensory stimuli. The question that The use of a passive ERP paradigm is not sufficient to
arises is how we can disentangle automatic from vol- reliably disentangle VS from MCS. Indeed, even if pas-
untary conscious brain activation. In 2006, Owen sive ERP paradigms can highlight ongoing brain pro-
et al. addressed this concern by applying an fMRI par- cessing for a given stimulus input, they do not
differentiate between automatic and voluntary cognitive
Downloaded by [UQ Library] at 23:29 04 November 2014

adigm in a traumatic VS patient who was asked to

perform two mental imagery tasks (“Imagine playing processes and, therefore, between unconscious and con-
tennis” and “Imaging visiting your house”). Activa- scious patients. For this reason, an active ERPs para-
tion was observed in the supplementary motor area digm was developed, where the participant is instructed
after the patient was asked to imagine playing tennis, to voluntarily direct attention to a target stimulus and to
and in parahippocampal gyrus when she was asked to ignore other stimuli (Figure 1) (Schnakers et al., 2008b,
imagine visiting her house. Similar activation patterns 2009a). Group as well as individual results showed that
were seen in healthy volunteers (Boly et al., 2007a). a larger P300 response to the own name was observed
Importantly, because the only difference between the in MCS patients in active condition as compared to pas-
conditions that elicited task-specific activation was in sive listening. This P300 amplitude was otherwise equi-
the instruction given at the beginning of each scan- valent to that observed in controls, while no task-related
ning session, the activation observed can only reflect P300 changes in VS patients were observed (Schnakers
the intentions of the patient, rather than some altered et al., 2008b). This suggests that MCS patients were
property of the outside world. Some could argue that able to voluntarily focus their attention on the target as a
the words “tennis” and “house” may have automati- function of task requirements. This active ERPs para-
cally triggered the patterns of activation observed in digm permitted to detect consciousness in a total LIS
target brain areas in this patient in the absence of con- patient (i.e., characterized by complete immobility
scious awareness. Although it is well documented that including all eye movements; Bauer, Gerstenbrand, &
these words elicit wholly automatic neural responses Rumpl, 1979) that behaviorally would be diagnosed as
in the absence of conscious awareness (Hauk, John- comatose (Schnakers et al., 2009a). In a similar way,
srude, & Pulvermuller, 2004), such responses typi- conscious processing was detected in three out of four
cally last for a few seconds and occur in regions of the MCS patients who were instructed to actively count the
brain that are associated with word processing. In this number of deviant trials in series of sound, while no
patient, the observed activity persisted for the full 30 s response were recorded in VS patients (Bekinschtein
of each imagery task and persisted until the patient et al., 2009).
was cued with another stimulus indicating that she
should rest. Thus, such responses cannot be explained
in terms of automatic brain processes (Soddu et al., RESTING STATE fMRI PARADIGM: A
2009). In addition, the responses in the patient were NEW TOOL TO CATEGORIZE
observed not in brain regions that are known to be DISORDERS OF CONSCIOUSNESS
involved in word processing but, rather, in regions PATIENTS?
that are known to be involved in the two imagery
tasks that she was asked to carry out (Owen et al., Resting state fMRI acquisitions are easy to perform and
2007). In this sense, the decision to “imagine playing could have a potentially broader and faster translation
tennis” rather than simply “rest” is an act of willed into clinical practice. Recent studies on spontaneous
intention and, therefore, clear evidence for response fluctuations in the fMRI blood-oxygen-level-dependent
to command awareness. The results of this study (BOLD) signal recorded in “resting” awake healthy
should not be misinterpreted as evidence that all subjects showed the presence of coherent fluctuations
 DISORDERS OF CONSCIOUSNESS 197
Downloaded by [UQ Library] at 23:29 04 November 2014

Figure 1. Illustration of active and passive event-related potential paradigms in a 22-year-old locked-in syndrome patient. Upper panel
depicts the patient’s response to an unfamiliar name (listened nontarget unfamiliar name, in blue) as compared with active condition
(responded to target unfamiliar name, in red). Lower panel depicts the patient’s own-name response in the passive condition (listened nontarget
own name, in green) and in the active condition (responded to target own name, in pink). (Adapted from data in Schnakers et al., 2009a.)

among functionally defined neuroanatomical networks 1 ischemia, 1 encephalopathy, 1 traumatic; and 1 hem-
(Boly et al., 2008b; Raichle, 2006). The concept of a orrhagic brain-dead patients (published in Boly et al.,
“default mode network” describes a set of brain areas 2009), age range 27–77 years, all men) to 9 other
exhibiting task-induced deactivations, encompassing healthy volunteers (age range 29–65 years, 5 women).
precuneus, posterior parietal lobe, and medial prefrontal In patients, clinical examination was repeatedly per-
cortex, which are more active at rest than when we are formed using standardized scales (the Coma Recovery
involved in attention-demanding cognitive tasks (Rai- Scale Revised, Giacino Kalmar, & Whyte, 2004b; and
chle et al., 2001). The clinical interest of default net- the Glasgow Liege scale, Born, 1988) on the day of
work MRI studies is that they allow the investigation of scanning, and in the week before and the week after.
higher order cognitive networks without requiring Patients were scanned in an unsedated condition. The
patients’ active participation, particularly important in study was approved by the Ethics Committee of the
VS and MCS. We here compared default mode network Medical School of the University of Liège. Informed
in VS and brain-dead patients with healthy volunteers. consent to participate to the study was obtained from
the subjects themselves in the case of healthy subjects,
and from the legal surrogates of the patients.
Patients

We first created a healthy control template of the Data acquisition and analysis
default mode network (DMN) by analyzing 11 volun-
teers (age range 21–60 years; 4 women). Secondly, In all participants, resting state BOLD data were
we compared 4 brain injured patients (3 vegetative: acquired on a 1.5 T MR scanner (Symphony Tim,
 198 BRUNO ET AL.

Siemens, Germany) with a gradient echo-planar the template identified by the black and white contour.
sequence using axial slice orientation (36 slices; voxel The principal brain regions characterizing the DMN,
size = 3.75 × 3.75 × 3.6 mm3; matrix size = 64 × 64 × posterior cingulate/precuneal, mesiofrontal and poste-
36; repetition time = 3000 ms; echo time = 30 ms; flip rior parietal cortices, were detected. Compared to con-
angle = 90°; field of view = 240 mm). A protocol of trols, VS patients showed a significant lower spatial
200 scans with a duration of 10 min was performed. similarity (0.47 ± 0.08, range 0.33–0.55 vs. 0.23 ± 0.12,
fMRI data were preprocessed using BrainVoyager soft- range 0.12–0.35; p = .05). The brain-dead patient (and
ware (R. Goebel, Brain Innovation, Maastricht, The the phantom) also showed lower spatial similarity com-
Netherlands). Preprocessing of functional scans pared to controls with values in the VS range. Finally,
included 3D motion correction, linear trend removal, none of the VS patients showed a DMN with a spatial
slice scan time correction, and filtering out of low fre- pattern comparable with healthy controls (Figure 2b–d),
quencies of up to 0.005 Hz. The data were spatially as assessed by visual inspection, even if in the case of
smoothed with a Gaussian filter of full width at half V1 the spatial similarity had a value consistent with
maximum value of 8 mm. Independent component ana- healthy subjects (Table 1). Rapid transient “clonic”
lysis (ICA) was performed with BrainVoyager using motions (i.e., spasmodic changes of muscular contrac-
30 components (Ylipaavalniemi & Vigario, 2008). An tion) were observed for VS1 and VS3 as also confirmed
average template DMN map was calculated on 11 by spatial maps’ periphery patterns. The brain-dead
Downloaded by [UQ Library] at 23:29 04 November 2014

healthy subjects. We performed, as implemented in patient didn’t show any significant spatial pattern, con-
BrainVoyager (self-organizing ICA; Esposito et al., firming results of Boly and collaborators (2009).
2005), a spatial similarity test on single subjects inde-
pendent components (IC) and we averaged the maps of
the ICs belonging to the same cluster. The cluster cor- DISCUSSION
responding to the DMN was selected by visual inspec-
tion. In order to select the DMN for each subject and After creating a template of the DMN, we identified
patient, we run self-organizing ICA with the average the DMN at the single subject level, in a user-inde-
ICs coming from the 11 healthy subjects template and pendent manner. In healthy volunteers, the identified
we picked the component clustering with the DMN component showed the typical spatial pattern of the
template (spatial similarity test). Self-organizing ICA DMN (Beckmann, DeLuca, Devlin, & Smith, 2005).
could assign a similarity value that indicates how well In VS patients, as well as in the brain-dead, we failed
the selected IC fitted the average DMN map based on to show any consistent pattern of DMN, even if single
the 11 subjects (the similarity test was based only on subjects brain activation maps showed residual con-
spatial properties without excluding components with a nectivity in VS. In our view, these residual connectiv-
power spectrum dominated by high frequency). The ity patterns are not reflecting residual DMN neuronal
same protocol of 200 scans was acquired on a spherical activity but could be explained by movement artifacts.
phantom. Finally, a two-tailed unequal variance Stu- Cardiorespiratory and vascular effects could also be a
dent t-test compared the spatial similarity values of the source of artifactual connectivity. BOLD signal
selected DMN map in healthy controls compared to VS changes within regions of the DMN have been found
patients. Three motion indices were introduced describ- to be reduced after correcting for cardiorespiratory
ing the motion of patients compared to healthy con- effects (van Buuren et al., 2009). In addition, it should
trols: the mean frequency calculated as the mean of the be stressed that altered blood flow and physiology,
frequency of each motion curve (three translations and such as neurovascular decoupling, could also effect
three rotations), the mean over time of the full displace- the BOLD signal detected in fMRI (Logothetis, Pauls,
ment during the acquisition calculated as the square Augath, Trinath, & Oeltermann, 2001), especially in
root of the sum of the squares of the six motion param- the context of traumatic and ischemic encephalopa-
eters, and the mean over time of the displacement thy. This of course cannot exclude the absence of neu-
speed during the acquisition calculated as the square ral activity in these patients (Boly et al., 2009; van
root of the sum of the squares of the six motion param- Buuren et al., 2009). Boly and collaborators (2009)
eters variation over one time unit (repetition time). showed that correlations with posterior cingulate
cortex are reduced in one VS as compared to age-
matched controls, and argued that this reduced
RESULTS functional connectivity within the DMN in VS was in
line with the hypothesis that spontaneous fMRI signal
A healthy controls mean spatial map identified the changes may be partly related to ongoing interoceptive
DMN pattern (Figure 2a) showing full overlap with states of mind or conscious thoughts.
 DISORDERS OF CONSCIOUSNESS 199
Downloaded by [UQ Library] at 23:29 04 November 2014

Figure 2. Resting state EPI–BOLD acquisition in (a) 9 healthy volunteers, with a spatial map obtained by running a random effect group
GLM analysis using as predictors the time courses of the ICs selected as default mode (thresholded at false discovery rate corrected p < .05)
within default mode mask obtained from an independent dataset (the 11 healthy controls for the independent study) shown as black and white
contour volume of interest. (b–d) 3 vegetative patients and (e) 1 brain death with spatial maps obtained by running a GLM analysis using as
predictor the time courses of the ICs selected as default mode (thresholded at false discovery rate corrected p < .05).
 200 BRUNO ET AL.

TABLE 1
Spatial and temporal properties of the independent component selected as default mode (DM) from resting state fMRI in healthy
controls, vegetative (VS) and one brain death (BD) patients (phantom is added for comparison)

Normalized DM time Motion

Spatial variation of course mean curves mean Mean
similarity spatial similarity frequency (Hz) frequency (Hz) displacement Mean speed

Control 1 0.33 1.5 0.04 0.03 0.6 0.04

Control 2 0.51 2.9 0.05 0.06 0.3 0.14
Control 3 0.37 1.8 0.05 0.03 1.7 0.09
Control 4 0.50 2.8 0.05 0.06 0.4 0.14
Control 5 0.54 3.2 0.07 0.05 0.8 0.25
Control 6 0.39 2 0.04 0.04 0.3 0.02
Control 7 0.55 3.2 0.05 0.03 3.0 0.09
Control 8 0.54 3.2 0.05 0.04 0.3 0.03
Control 9 0.47 2.6 0.04 0.03 0.8 0.05
VS 1 0.35 1.7 0.04 0.05 1.1 0.14
VS 2 0.23 0.8 0.07 0.05 0.2 0.28
VS 3 0.12 –0.1 0.04 0.03 5.9 0.50
Downloaded by [UQ Library] at 23:29 04 November 2014

BD 0.13 0 0.07 0.06 0.1 0.27

Phantom 0.12 –0.1 0.06 0.03 0.1 0.24

Notes: Spatial similarity with an average-template-default mode based on an independent 11 healthy controls data set, normalized variation of
spatial similarity (Sim) respect to the brain death (BD) (Sim_subject-Sim_BD)/Sim_BD), and default mode time course mean frequency. Motion
properties (means calculated from the six motion curves). Rapid transient “clonic” motions were observed for two of the three patients (VS1 and
VS3) as also confirmed by spatial maps periphery patterns, while a drift (i.e., a displacement with low speed) is observed for control 3 and 7.

Motion, pulse, and respiratory artifacts remain a very diagnosed by means of exclusion criteria such as “no
important problem to tackle to be able to properly assess goal-directed eye movements” or “no execution of com-
patients with disorders of consciousness. If recording mands.” Those behavioral signs may be very slight and
heart and respiratory rates during the acquisition can short-term and performance may fluctuate. Integration of
help in disentangling neuronal activity from pulse and neuroimaging and ERPs techniques should improve our
respiratory artifacts (Gray et al., 2009), further investi- ability to disentangle diagnostic and prognostic differ-
gations are needed to disentangle effects of motion and ences on the basis of underlying mechanisms and better
neuronal activity in the BOLD signal. In the brain-dead guide our clinical therapeutic options in these challeng-
patient, who fulfilled all the standard clinical criteria for ing patients. A step-by-step approach combining multi-
brain death (previously published in Boly et al., 2009), modal assessment techniques (i.e., PET scan, fMRI, and
fMRI results did not show any long-range significant ERPs) seems to be appropriate to detect signs of con-
functional connectivity, confirming that the origin of the sciousness (Figure 3). Resting metabolism and passive
BOLD signal was fully due to motion artifacts as paradigms studies increased our understanding of resid-
expected due to the complete absence of neuronal activ- ual cerebral processing of VS and MSC patients. Active
ity in brain death. Finally, future work should study the paradigms seem to provide an objective valuable addi-
relationship between structural and functional connec- tional diagnostic tool in cases of patients with atypical
tivity changes in noncommunicative brain-damaged activation, leading to persisting doubts in clinical diag-
patients. Functional connectivity in DMN has indeed nosis. Negative results, however, must be cautiously
been related to underlying structural anatomy (Greicius, interpreted in case of patients with severely altered level
Supekar, Menon, & Dougherty, 2009). Further multi- of vigilance, which could present only transient activity
modal studies should combine resting state fMRI data in response to the presentation of instructions. Passive
with structural MRI (e.g. diffusion tensor imaging data). and active fMRI paradigms were also shown to poten-
tially be a useful tool to predict possibility of recovery.
Finally, results on the DMN seem promising to dis-
CONCLUSION tinguish patients with disorders of consciousness by
evaluating preservation of connectivity in this resting
Patients with disorders of consciousness represent a state network. However, future studies are needed to
major clinical problem in terms of clinical assessment, give a full characterization of default mode connectivity
treatment, and daily management. The nonresponsive- in VS and MCS patients and its potential use in outcome
ness of such patients implies that they can only be prediction. Future clinical studies should also perform
 DISORDERS OF CONSCIOUSNESS 201

American Congress of Rehabilitation Medicine (1995).

Recommendations for use of uniform nomenclature
pertinent to patients with severe alterations of conscious-
ness. Archives of Physical Medicine and Rehabilation,
76, 205–209.
Bauer, G., Gerstenbrand, F., & Rumpl, E. (1979). Varieties
of the locked-in syndrome. Journal of Neurology,
221(2), 77–91.
Beckmann, C. F., DeLuca, M., Devlin, J. T., & Smith, S. M.,
(2005). Investigations into resting-state connectivity
using independent component analysis. Philosophical
Transactions of the Royal Society of London B: Biological
Sciences, 360, 1001–1013.
Bekinschtein, T., Niklison, J., Sigman, L., Manes, F.,
Leiguarda, R., Armony, J., et al. (2004). Emotion pro-
cessing in the minimally conscious state. Journal of
Neurology, Neurosurgery & Psychiatry, 75(5), 788.
Bekinschtein, T., Dehaene, S., Rohaut, B., Tadel, F., Cohen,
L., & Naccache, L. (2009). Neural signature of the con-
scious processing of auditory regularities. Proceedings
Downloaded by [UQ Library] at 23:29 04 November 2014

of the National Academy of Sciences of the United States

of America, 106(5), 1672–1677.
Boly, M., Coleman, M. R., Davis, M. H., Hampshire, A.,
Bor, D., Moonen, G., et al. (2007a). When thoughts
become action: An fMRI paradigm to study volitional
brain activity in non-communicative brain injured
patients. NeuroImage, 36(3), 979–992.
Boly, M., Faymonville, M. E., Peigneux, P., Lambermont,
B., Damas, P., Del Fiore, G., et al. (2004). Auditory
processing in severely brain injured patients: Differ-
ences between the minimally conscious state and the
persistent vegetative state. Archives of Neurology,
61(2), 233–238.
Boly, M., Faymonville, M. E., Schnakers, C., Peigneux, P.,
Lambermont, B., Phillips, C., et al. (2008a). Perception
of pain in the minimally conscious state with PET acti-
vation: An observational study. Lancet Neurology,
7(11), 1013–1020.
Boly, M., Phillips, C., Tshibanda, L., Vanhaudenhuyse,
A., Schabus, M., Dang-Vu, T. T., et al. (2008b). Intrin-
sic brain activity in altered states of consciousness:
Figure 3. Illustration of a step-by-step approach combining PET How conscious is the default mode of brain function?
scan, ERPs, and fMRI techniques which might reveal signs of Annals of the New York Academy of Sciences, 1129,
consciousness that are unattainable by bedside clinical assessment 119–129.
of disorders of consciousness patients. These high-tech devices Boly, M., Tshibanda, L., Vanhaudenhuyse, A., Noirhomme,
might permit some of these patients to show their consciousness by Q., Schnakers, C., Ledoux, D., et al. (2009). Functional
following commands (e.g., to imagine playing tennis, respond to connectivity in the default network during resting state is
own name) via nonmotor pathways. preserved in a vegetative but not in a brain dead patient.
Human Brain Mapping, 30(8), 2393–2400.
Born, J. D. (1988). The Glasgow-Liège Scale: Prognostic
value and evaluation of motor response and brain stem
heart rate and respiratory rate recording and simultane- reflexes after severe head injury. Acta Neurochirurgica,
ous real movement monitoring, which would improve 95, 49–52.
default mode analyses. Bruno, M. A., Schnakers, C., Damas, F., Pellas, F., Lutte, I.,
Bernheim, J., et al. (2009). Locked-in syndrome in chil-
dren: Report of five cases and review of the literature.
Pediatric Neurology, 41(4), 237–246.
REFERENCES Coleman, M. R., Rodd, J. M., Davis, M. H., Johnsrude, I. S.,
Menon, D. K., Pickard, J. D., et al. (2007). Do vegetative
Alkire, M. T., Pomfrett, C. J., Haier, R. J., Gianzero, M. V., patients retain aspects of language comprehension?
Chan, C. M., Jacobsen, B. P., et al. (1999). Functional Evidence from fMRI. Brain, 130(10), 2494–2507.
brain imaging during anesthesia in humans: Effects of Dehaene, S., & Naccache, L. (2001). Towards a cognitive
halothane on global and regional cerebral glucose neuroscience of consciousness: Basic evidence and a
metabolism. Anesthesiology, 90(3), 701–709. workspace framework. Cognition, 79(1–2), 1–37.
 202 BRUNO ET AL.

Demertzi, A., Schnakers, C., Ledoux, D., Chatelle, C., lamocortical connectivity after recovery from persistent
Bruno, M. A., Vanhaudenhuyse, A., et al. (2009). Differ- vegetative state. Lancet, 355(9217), 1790–1791.
ent beliefs about pain perception in the vegetative and Laureys, S., Faymonville, M. E., Moonen, G., Luxen, A., &
minimally conscious states: A European survey of med- Maquet, P. (2000c). PET scanning and neuronal loss in
ical and paramedical professionals. Progress in Brain acute vegetative state. Lancet, 355, 1825–1826.
Research, 177, 329–338. Laureys, S., Goldman, S., Phillips, C., Van Bogaert, P.,
Di, H. B., Yu, S. M., Weng, X. C., Laureys, S., Yu, D., Li, Aerts, J., Luxen, A., et al. (1999a). Impaired effective
J. Q., et al. (2007). Cerebral response to patient’s own cortical connectivity in vegetative state. NeuroImage,
name in the vegetative and minimally conscious states. 9(4), 377–382.
Neurology, 68(12), 895–899. Laureys, S., Lemaire, C., Maquet, P., Phillips, C., & Franck,
Esposito, F., Scarabino, T., Hyvarinen, A., Himberg, J., G. (1999b). Cerebral metabolism during vegetative state
Formisano, E., Comani, S., et al. (2005). Independent and after recovery to consciousness. Journal of Neurology,
component analysis of fMRI group studies by self- Neurosurgery & Psychiatry, 67(1), 121.
organizing clustering. NeuroImage, 25, 193–205. Laureys, S., Owen, A. M., & Schiff, N. D. (2004a). Brain
Fernandez-Espejo, D., Junque, C., Vendrell, P., Bernabeu, function in coma, vegetative state, and related disorders.
M., Roig, T., Bargallo, N., et al. (2008). Cerebral Lancet Neurology, 3(9), 537–546.
response to speech in vegetative and minimally con- Laureys, S., Perrin, F., Faymonville, M. E., Schnakers, C.,
scious states after traumatic brain injury. Brain Injury, Boly, M., Bartsch, V., et al. (2004b). Cerebral processing in
22(11), 882–890. the minimally conscious state. Neurology, 63(5), 916–918.
Giacino, J. T. (2004a). The vegetative and minimally con- Levy, D. E., Sidtis, J. J., Rottenberg, D. A., Jarden, J. O.,
Downloaded by [UQ Library] at 23:29 04 November 2014

scious states: Consensus-based criteria for establishing Strother, S. C., Dhawan, V., et al. (1987). Differences in cer-
diagnosis and prognosis. NeuroRehabilitation, 19(4), ebral blood flow and glucose utilization in vegetative versus
293–298. locked-in patients. Annals of Neurology, 22(6), 673–682.
Giacino, J. T., Kalmar, K., & Whyte, J. (2004b). The JFK Logothetis, N. K., Pauls, J., Augath, M., Trinath, T., & Oel-
Coma Recovery Scale – Revised: Measurement charac- termann, A. (2001). Neurophysiological investigation of
teristics and diagnostic utility. Archives of Physical and the basis of the fMRI signal. Nature, 412, 150–157.
Medical Rehabilitation, 85(12), 2020–2029. Maquet, P., Dive, D., Salmon, E., Sadzot, B., Franco, G.,
Giacino, J. T., Ashwal, S., Childs, N., Cranford, R., Jennett, Poirrier, R., et al. (1990). Cerebral glucose utilization
B., Katz, D. I., et al. (2002). The minimally conscious during sleep–wake cycle in man determined by positron
state: Definition and diagnostic criteria. Neurology, emission tomography and [18F]2-fluoro-2-deoxy-D-glu-
58(3), 349–353. cose method. Brain Research, 513(1), 136–143.
Gray, M. A., Minati, L., Harrison, N. A., Gianaros, P. J., Menon, D. K., Owen, A. M., Williams, E. J., Minhas, P. S.,
Napadow, V., & Critchley, H. D. (2009). Physiological Allen, C. M., Boniface, S. J., et al. (1998). Cortical pro-
recordings: Basic concepts and implementation during cessing in persistent vegetative state. Lancet, 352(9123),
functional magnetic resonance imaging. NeuroImage, 200.
47, 1105–1115. Moritz, C. H., Rowley, H. A., Haughton, V. M., Swartz, K.
Greicius, M. D., Supekar, K., Menon, V., & Dougherty, R. F. R., Jones, J., & Badie, B. (2001). Functional MR imag-
(2009). Resting-state functional connectivity reflects ing assessment of a non-responsive brain injured patient.
structural connectivity in the default mode network. Magnetic Resonance Imaging, 19(8), 1129–1132.
Cerebral Cortex, 19, 72–78. Multi-Society Task Force on PVS, The (1994). Medical
Hauk, O., Johnsrude, I., & Pulvermuller, F. (2004). Somato- aspects of the persistent vegetative state (1). New Eng-
topic representation of action words in human motor and land Journal of Medicine, 330(21), 1499–1508.
premotor cortex. Neuron, 41(2), 301–307. Nakayama, N., Okumura, A., Shinoda, J., Nakashima, T., &
Kwong, K. K., Belliveau, J. W., Chesler, D. A., Goldberg, Iwama, T. (2006). Relationship between regional cere-
I. E., Weisskoff, R. M., Poncelet, B. P., et al. (1992). bral metabolism and consciousness disturbance in trau-
Dynamic magnetic resonance imaging of human brain matic diffuse brain injury without large focal lesions: An
activity during primary sensory stimulation. Proceed- FDG–PET study with statistical parametric mapping
ings of the National Academy of Sciences of the United analysis. Journal of Neurology, Neurosurgery & Psychi-
States of America, 89, 5675–5679. atry, 77(7), 856–862.
Laureys, S. (2005). The neural correlate of (un)awareness: Onofrj, M., Thomas, A., Paci, C., Scesi, M., & Tombari, R.
Lessons from the vegetative state. Trends in Cognitive (1997). Event related potentials recorded in patients with
Sciences, 9(12), 556–559. locked-in syndrome. Journal of Neurology, Neurosur-
Laureys, S., Berré, J., & Goldman, S. (2001). Cerebral gery & Psychiatry, 63(6), 759–764.
function in coma, vegetative state, minimally Owen, A., Coleman, M., Boly, M., Davis, M., Laureys, S.,
conscious state, locked-in syndrome and brain death. Jolles, J., et al. (2007). Response to comments on
In J. L. Vincent (Ed.), 2001 Yearbook of intensive care “Detecting awareness in the vegetative state”. Science,
and emergency medicine. (pp. 386–396). Berlin: 315(5816), 1221.
Springer-Verlag. Owen, A. M., Coleman, M. R., Boly, M., Davis, M. H.,
Laureys, S., Faymonville, M. E., Degueldre, C., Fiore, G. Laureys, S., & Pickard, J. D. (2006). Detecting aware-
D., Damas, P., Lambermont, B., et al. (2000a). Auditory ness in the vegetative state. Science, 313(5792), 1402.
processing in the vegetative state. Brain, 123(8), 1589– Owen, A. M., Menon, D. K., Johnsrude, I. S., Bor, D., Scott,
1601. S. K., Manly, T., et al. (2002). Detecting residual cogni-
Laureys, S., Faymonville, M. E., Luxen, A., Lamy, M., tive function in persistent vegetative state. Neurocase,
Franck, G., & Maquet, P. (2000b). Restoration of tha- 8(5), 394–403.
 DISORDERS OF CONSCIOUSNESS 203

Perrin, F., Schnakers, C., Schabus, M., Degueldre, C., state: Clinical consensus versus standardized neurobe-
Goldman, S., Bredart, S., et al. (2006). Brain response havioral assessment. BMC Neurology, 9, 35.
to one’s own name in vegetative state, minimally con- Smart, C. M., Giacino, J. T., Cullen, T., Moreno, D. R.,
scious state, and locked-in syndrome. Archives of Neu- Hirsch, J., Schiff, N. D., et al. (2008). A case of locked-in
rology, 63(4), 562–569. syndrome complicated by central deafness. Nature Clinical
Posner, J., Saper, C., Schiff, N., & Plum, F. (2007). Plum Practice Neurology, 4(8), 448–453.
and Posner’s diagnosis of stupor and coma. New York: Soddu, A., Boly, M., Nir, Y., Noirhomme, Q., Vanhauden-
Oxford University Press. huyse, A., Demertzi, A., et al. (2009). Reaching across
Raichle, M. E. (2006). Neuroscience: The brain’s dark the abyss: Recent advances in functional magnetic reso-
energy. Science, 314(5803), 1249–1250. nance imaging and their potential relevance to disorders
Raichle, M. E., MacLeod, A. M., Snyder, A. Z., Powers, W. J., of consciousness. Progress in Brain Research, 177,
Gusnard, D. A., & Shulman, G. L. (2001). A default mode 261–74.
of brain function. Proceedings of the National Academy of Staffen, W., Kronbichler, M., Aichhorn, M., Mair, A., &
Sciences of the United States of America, 98(2), 676–682. Ladurner, G. (2006). Selective brain activity in response
Rudolf, J., Ghaemi, M., Haupt, W. F., Szelies, B., & Heiss, to one’s own name in the persistent vegetative state.
W. D. (1999). Cerebral glucose metabolism in acute and Journal of Neurology, Neurosurgery & Psychiatry,
persistent vegetative state. Journal of Neurosurgical 77(12), 1383–1384.
Anesthesiology, 11(1), 17–24. Tommasino, C., Grana, C., Lucignani, G., Torri, G., &
Schiff, N. D., Rodriguez-Moreno, D., Kamal, A., Kim, K. Fazio, F. (1995). Regional cerebral metabolism of
H., Giacino, J. T., Plum, F., et al. (2005). fMRI reveals glucose in comatose and vegetative state patients. Jour-
Downloaded by [UQ Library] at 23:29 04 November 2014

large-scale network activation in minimally conscious nal of Neurosurgical Anesthesiology, 7(2), 109–116.
patients. Neurology, 64(3), 514–523. van Buuren, M., Gladwin, T. E., Zandbelt, B. B., van den
Schnakers, C., Majerus, S., Goldman, S., Boly, M., Van Heuvel, M., Ramsey, N. F., Kahn, R. S., et al. (2009).
Eeckhout, P., Gay, S., et al. (2008a). Cognitive function Cardiorespiratory effects on default-mode network
in the locked-in syndrome. Journal of Neurology, activity as measured with fMRI. Human Brain Mapping,
255(3), 323–330. 30(9), 3031–3042.
Schnakers, C., Perrin, F., Schabus, M., Hustinx, R., Maje- Vanhaudenhuyse, A., Laureys, S., & Perrin, F. (2008a).
rus, S., Moonen, G., et al. (2009a). Detecting conscious- Cognitive event-related potentials in comatose and post-
ness in a total locked-in syndrome: An active event- comatose states. Neurocritical Care, 8(2), 262–270.
related paradigm. Neurocase, 15, 271–277. Vanhaudenhuyse, A., Schnakers, C., Bredart, S., & Laureys,
Schnakers, C., Perrin, F., Schabus, M., Majerus, S., Ledoux, S. (2008b). Assessment of visual pursuit in post-coma-
D., Damas, P., et al. (2008b). Voluntary brain processing tose states: Use a mirror. Journal of Neurology, Neuro-
in disorders of consciousness. Neurology, 71, 1614–1620. surgery & Psychiatry, 79(2), 223.
Schnakers, C., Vanhaudenhuyse, A., Giacino, J., Ventura, Ylipaavalniemi, J., & Vigario, R. (2008). Analyzing con-
M., Boly, M., Majerus, S., et al. (2009b). Diagnostic sistency of independent components: An fMRI illustra-
accuracy of the vegetative and minimally conscious tion. NeuroImage 39, 169–180.