Closed-loop tDCS in Patients in Minimally Conscious State

This research will test a closed-loop system using EEG-arousal measures (spectral entropy) to define the best moment of the day for application of transcranial direct current stimulation (tDCS) in patients in MCS This study aims at answering the following questions: 1. Is tDCS applied during high vigilance states more effective in increasing the level of conscious awareness than low vigilance states in patients in minimally conscious state (MCS)? 2. Is the EEG pattern (connectivity, complexity) different after application of active or sham tDCS at high vigilance or low vigilance states? 3. Is there a difference in the profile of tDCS-responders as compared to non-responders with regards to etiology, clinical diagnosis (MCS+/MCS-), age, gender, time post-injury, functional outcome, structural and functional neuroimaging findings and EEG markers?

During the last 20 years, with the rediscovering and development of noninvasive brain stimulation (NIBS); techniques such as transcranial direct current stimulation (tDCS) and transcranial magnetic stimulation (TMS) have flourished thanks to the advancements in functional imaging, elegant neurophysiological assessments, computer generated modeling, and most importantly, the groundwork of well-designed research methodology. TDCS represents a safe, inexpensive and straightforward technique that could be easily integrated in rehabilitation programs. In a first sham-controlled double-blind randomized crossover study, the effect of a single prefrontal tDCS was evaluated in a heterogeneous population of patients with disorders of consciousness (DOC), vegetative state (VS) and minimally conscious state (MCS), acute-subacute (\< 3months) and chronic, with traumatic or non-traumatic etiologies. At the individual level, tDCS responders were defined as patients who presented a new sign of consciousness (e.g., command following; visual pursuit; recognition, manipulation or localization of objects); after the active tDCS session, that was not present before nor during the sham tDCS session. 13/30 patients in MCS showed a tDCS-related improvement. 2 acute (\<3 months) patients in VS out of 25 showed a tDCS response (i.e. showed command following and visual pursuit present after the anodal stimulation not present at baseline or pre- or post- sham-tDCS). At group level, a treatment effect, as measured by the Coma Recovery Scale-Revised (CRS-R) was observed in the MCS but not in the VS patients' group. In addition, no tDCS related side effects were observed. In another sham-controlled study tDCS was applied for five days either over the primary motor cortex or over the left prefrontal cortex in 10 chronic patients with DOC. In this trial, the effect of tDCS have been assessed up to 12 months post stimulation. This study highlighted that even chronic patients in DOC could improve and the effect could last up to 12 months post tDCS. Several tDCS studies have been performed by the Coma Science Group (Dr Thibaut) and other groups. So far, a total of 221 patients with DOC have been included in eight studies without any serious adverse event. In addition, three other trials performed by Naro et al including a total of 59, did not report any adverse event either. Therefore, tDCS appears as a safe technique, especially when applied by experts and highly trained investigators with a strong background in both non-invasive brain stimulation and patients with disorders of consciousness. From a neurophysiological point of view, tDCS increases the neuronal excitability by facilitating the action potential release and modification of the excitability of NMDA receptors. Moreover, tDCS may strengthen task-related dynamical synaptic connections. The effect of a single tDCS session last about 60 to 90 min. While, when the stimulation is repeated for 10 to 20 sessions, the effects have been found to last up to 3 months after the end of the stimulation sessions. A major challenge in the application of tDCS to patients with DOC is the variability in the individual behavioral response. Indeed, in previous studies conducted by the Coma Science Group and other groups, the rate of responders was inconsistent across studies. As stated above, a responder is a patient who showed a new sign of consciousness following the application of active tDCS that was never observed beforehand nor before or after sham stimulation. However, given the specificity of crossover designs used in these studies, this definition was recently updated as a patient who showed a new sign of consciousness for the first time after the application of active tDCS. The next steps to optimize the applications of tDCS include a better identification of tDCS responders beforehand. To that end, Thibaut and colleagues conducted a retrospective study comparing the structural and metabolic neuroimagery profiles of patients who responded to tDCS (n=8) and patients who did not (n=13), by the means of structural MRI T1 data and 18-fluorodeoxyglucose position emission tomography. The results showed a greater atrophy in non-responders in regions including the left DLPFC, the medial-prefrontal cortex and the left thalamus as compared to responders. The same areas as well as the thalamus were hypometabolic in non-responders as compared to responders. This suggest that response to tDCS requires at least partial structural and metabolic preservation of the stimulated area, as well as in subcortical brains areas involved in attention and working memory. Later, the authors used the same sample and retrospectively analyzed resting state EEG brain connectivity. The results showed higher theta centrality in responders (as compared to non-responders) meaning this new biomarker can be used to predict tDCS response in patients with DOC (Thibaut et al. 2018). Another key component to responsiveness is the timing of the stimulations. Indeed, patients with DOC typically present fluctuations in vigilance impacting their responsiveness to any kind of external stimuli. Patients in MCS even show a periodicity of 70 minutes in these fluctuations (range 57-80), comparable to the fluctuations in attention observed in healthy controls, while patient in VS do not present this type of periodicity. Piarulli and colleagues used the spectral entropy measured with resting EEG to highlight the periodicity in its fluctuations and suggested that the EEG spectral entropy variability in MCS could mirror the fluctuation of awareness previously described in this population. Administering tDCS during specific time windows (i.e., periods of low or high arousal) could therefore influence its clinical efficacy in patients in MCS since it is known that the positive effects of tDCS are dependent on the brain neural state. To this end, recent advances in information technology enable the implementation of a closed-loop set-up by complex computations being performed in real-time. Using this technology to target specific levels of vigilance with tDCS in patient in MCS could provide insight in understanding the patterns of responding to that treatment and optimize future applications.