We have studied the electrophysiological evolution, with the aid of lo
ng-term, quantitative E.E.G. (Holter method) and sensory evoked potent
ials (visual, auditory and somatosensory) on 25 severe traumatic head
injury patients (initial Glasgow scale - between 3 and 5) aged from 5
to 42 years having evolved towards a persistent vegetative state and b
eing object of a protocol of thalamic stimulation. Electrophysiologica
l exploration was practised before placement of the stimulation, then
at 2 weeks, 1 month, 2 months and finally 3 months after. The electrop
hysiological parameters appear to have no predictive value on the clin
ical evolution after stimulation. However, changes such as long-term q
uantitative EEG as well as evoked potentials mostly precede-clinical p
rogress when present. Cortical evoked potentials are the first to chan
ge significantly (J15) followed by long-term quantitative EEG and brai
n stem evoked potentials (J30); the latter improving more completely.
Also the consideration of the chronobiological aspects of the recordin
gs of the long term EEG allow the proposal of <<waking targeted>> stim
ulations at the most favorable moments of the nyctohemeral period.