The fatal threshold

This method, which is based on the induced perturbation of a system, the brain, to calculate its ability to integrate information thanks to sophisticated algorithms, was elaborated by Marcello Massimini, professor at the University of Milan and visiting professor at the University of Liège. By applying it to subjects plunged into a deep sleep, the Italian researcher showed that this state induced a loss of brain connectivity. But what happened in anaesthetised subjects, people who were dreaming (paradoxical sleep) or severely brain-damaged patients presenting one of the following diagnoses: coma, vegetative state/unresponsive wakefulness syndrome, minimal consciousness or locked-in syndrome? This was the question raised by Steven Laureys’ team. It was the core element of an experimental study undertaken at Liège, during which 200 measurements were taken among 50 subjects, some awake, others asleep or anaesthetised, as well as severely brain-damaged patients.

“The principle behind this is that the more the brain generates information as an integrated whole, the less it is possible to compress its response when it is subjected to a strong stimulus. We assess the complexity of this response according to the electric echo generated", says Olivia Gosseries, FNRS researcher in the Coma Science Group and first author, alongside Adenauer Casali (Milan), of the article recently published in Science Translational Medicine.

The international group revealed that the measurement of complexity calculated on the basis of the cortical responses to transcranial magnetic stimulation, could reliably determine the level of consciousness in the subjects involved in the test. This measurement, which they named the Perturbational Complexity Index (PCI), oscillates between two extremes: 0 and 0.7. It became clearly apparent that there was a line (0.3) defining the presence or lack of consciousness. In people who are asleep or anaesthetised, or in patients in a coma or a vegetative/unresponsive state, the PCI is situated below the fatal threshold (0.3). On the other hand, this threshold is exceeded in patients in a minimally conscious state, those with locked-in syndrome as well as in wakeful control subjects, and the figure obtained seems to be proportional to the level of consciousness. In LIS patients, for instance, it is normal (between 0.5 and 0.7), whereas it varies between 0.3 and 0.5 in a minimally conscious state. “We haven’t seen any errors or any exceptions to the rule”, specifies Steven Laureys. We can therefore consider that, for the first time, the researchers have succeeded in creating an objective measuring scale covering the conscious-unconscious spectrum.

Tracking down consciousness

With the aim of validating the technique, the brain-damaged patients selected for this first study had obviously received a diagnosis with a very high degree of certainty. Ambiguous cases, where it wasn’t possible to determine whether the patient was in a vegetative/unresponsive state or in a minimally conscious state, weren’t included. However, it is exactly these cases that the researchers are going to focus on today. Because, from a clinical point of view, they are at the centre of the debate. It is for these people that a measurement tool such as the PCI could play a determining role by almost incontestably specifying whether or not they have residual consciousness.