How do we detect signs of consciousness in severely brain-injured patients and, subsequently, make the right diagnosis? Or, how can we communicate with people with locked-in syndrome (LIS)? Of course, there are imagining systems that obtain their information directly from the brain, but their error rates remain significant, they are expensive and unwieldy. Consequently, for several years now, the focus has been on designing tools that are sensitive, inexpensive and easy to use. The University of Liège’s Coma Science Group has taken part in various projects of this type, with the primary focus being brain-computer interfaces (BCIs). One of the most successful of these is the subject of Damien Lesenfants’ doctoral thesis.  Its basic principle? One or several stimuli oscillating at constant and different frequencies are presented to the patient. When the latter focuses their attention on a stimulus, an increase in electroencephalographic activity at the frequency of the stimulus is detected in the posterior regions of the brain, particularly in the occipital areas.

An increasing number of lives are saved thanks to progress in resuscitative medicine. However, alongside these successes is another reality: the rise in the number of brain-injured patients plunged into an irreparable altered state of consciousness (vegetative / unresponsive state, minimally conscious state), or even locked-in syndrome (LIS), where the patient is aware but can’t move.

The limit between a vegetative/unresponsive state and a minimally conscious state is difficult to define through a clinical assessment of the patient. And often, LIS patients are wrongly thought to be in one of these states. Traditionally, the clinical assessment of patients is based on the examination of their motor responses using behavioural scales - for instance, can they squeeze their hand when asked to, or follow a moving object with their eyes? As shown by Caroline Schnakers, an FNRS (National Funds for Scientific Research) post-doctoral researcher, and Professor Steven Laureys, director of the Coma Science Group (University of Liège/Liège CHU), the diagnosis is wrong in 40 % of cases.

In his recently-defended thesis, which he wrote as a member of the Coma Science Group, Damien Lesenfants, a civil engineer specialising in biomedical research, wrote: “The behavioural scales are mainly based on the subject’s motor responses and verbal understanding. This makes the diagnosis difficult among this population which often suffers from motor disorders, aphasia and fluctuating vigilance. Furthermore, the reliability of the diagnosis with these tools depends on the investigator’s experience. Consequently, it is important to develop paramedical tools that are objective and independent from the motor tests, in order to detect these signs of consciousness when no obvious response has been observed at the patient’s bedside.”

Cognitive event-related potentials

Systematic recourse to a standardised and sensitive behavioural scale, such as the Coma Recovery Scale-Revised - CRS-R, developed in the United States by Joseph Giacino’s team and validated in French and Dutch by Caroline Schnakers and Steven Laureys, has brought the diagnosis error rate down to 31 %. And when neurologists use functional magnetic resonance imaging (fMRI) to fine tune the latter, or better still, positron emission tomography (PET scan), techniques which allow us to indirectly "see" the brain in action, the error rate falls to around 20 %. "fMRI and PET scans obtain the information directly from the brain”, Damien Lesenfants explains. “Therefore, they are likely to show a possible activation in response to a command even if the patient, who is suffering from motor disorders, can’t move.” However, the residual error rate is still quite high. Moreover, fMRI and PET scans have several disadvantages, notably their cost, lack of availability, their non-portability, their sensitivity to the patient’s movements – if they move, the information collected may be rendered useless – and the time it takes to acquire the data, since the people being examined have a very limited capacity to concentrate.

(1) Lesenfants, D., "Interface cerveau-ordinateur, locked-in syndrome et troubles de la conscience", septembre 201