Conference Agenda

Neural field theory offers a robust framework for understanding how the activation of neuronal populations effects large-scale electrophysiological (EEG) activity based on physiologically measurable parameters. This provides insights into the mechanisms underlying observable EEG features. Thalamocortical loops have been proposed to be crucial for the generation of consciousness. In particular, disorders of consciousness (DoC) often involve disrupted feedforward/feedback interactions within this circuit. In this study we applied a physiological neural field model, previously validated for reproducing EEG spectral characteristics in healthy subjects, to analyse EEG spectra in DoC patients and healthy controls. The model includes four neuronal populations: cortical excitatory and inhibitory, and thalamic relay and reticular nuclei, and the connections between them.

Using data from two independent DoC patient cohorts in Liège and Paris, we fitted EEG power spectra from 275 patients with unresponsive wakefulness syndrome (UWS), 332 in the minimally conscious state (MCS), and 70 healthy controls. Thalamocortical loop gains were significantly impaired in DoC patients, differentiating UWS and MCS. UWS patients exhibited shorter delays between the thalamus and cortex and lower gains in excitatory-inhibitory loops, aligning with the mesocircuit hypothesis and highlighting the thalamus's critical role in sustaining consciousness. Furthermore, spectral model parameters were inverted to the time domain, this successfully replicated EEG time series evidenced by the preserved complexity differences between groups. These findings support the model's potential for studying the mechanisms of the reported complexity in brain signals and its possible applicability to virtual treatment trials aimed at restoring consciousness.

The clinical interpretation of eyes-opening/closing after severe brain injury remains elusive. The straightforward relationship between the wakefulness and awareness can be unreliable in altered states of consciousness caused by such injury. We hypothesise that a persistent coupling between brain connectivity and behaviour anticipates a favourable clinical outcome.

We evaluated the relationship between fluctuations in eyes-opening/closing and those in brain connectivity using 24h-recordings acquired from 18 patients with acute disorders of consciousness. We analysed 13-channel EEG-recordings grouped into 5 regions of interest (ROI: frontal, central, parietal, temporal, occipital). Brain connectivity (weighted Phase Lag Index) was calculated between 10 combinations of ROIs within 5 spectral frequency bands (Delta, Theta, Alpha, Sigma, Beta). These connectivity time courses were correlated to periods of eyes opening/closing using video recordings.

We observed significant correlations between connectivity and eyes open/closed periods (Spearman rho-values after Bonferroni correction). We then measured the relationship between the dichotomized final clinical outcome (favourable if patients reached the Exit-MCS group) and the significant wPLI/eyes opening correlation (Kruskal-Wallis statistic).

We observed that patients with the favourable clinical outcomes had a stronger correlation between brain connectivity and eye-opening, notably when correlations were negative, occurred in the Beta band and over long-range connectivity spans (Frontal<->Central/Parietal/Occipital; Central<->Occipital/Parietal).

Our findings demonstrate that eyes-opening reappearance can have radically different implications depending on its neurophysiological correlate. Such connectivity changes implies that some patients would have more than a sub-cortically driven eye-opening behaviour: it confirms cortical function and creates a new sign of cortical function among MCS and UWS patients, respectively.

Over the past fifty years, clinicians and researchers have developed different strategies to detect recovery of consciousness after severe brain injury. These approaches can be categorized into four hierarchical levels, respectively probing: (1) overt behavior, assessed with tools like the Coma Recovery Scale Revised (CRS-R); (2) voluntary neural activations, measured through active fMRI/EEG paradigms; (3) perceptual neural responses, such as the P3 in oddball paradigms; and (4) intrinsic brain properties, like resting EEG spectral properties and measures of perturbational complexity (i.e. Perturbational Complexity Index - PCI). These tools variably rely on intact sensory, motor, and executive functions, which are often impaired after brain injury, potentially affecting their sensitivity in detecting consciousness recovery.

We systematically compared these methodologies in a selected cohort of 12 patients admitted to an Intensive Rehabilitation Unit and exhibiting prolonged unresponsiveness (persisting for 5 to 25 weeks post-injury) as per CRS-R assessments, despite no major negative prognostic factors in intensive care. Eventually, all recovered behavioral signs of consciousness (MCS- or beyond). During the unresponsive phase, 25% of patients ranked positive in EEG active paradigms, 20% in clinical P3-like component, 83% in EEG spectral decay, and 100% in PCI. Clinical evoked potentials and structural imaging revealed sensory-motor deafferentation in all cases along with instances of frontal executive system damage, potentially explaining an earlier detection of consciousness recovery by measures solely relying on intrinsic brain properties.

These findings underscore the utility of deeper layers of investigation in patients showing prolonged unresponsiveness in the subacute phase after severe brain injury.

Background/aims: The neurochemical deficits underlying disorders of consciousness (DoC) are largely unknown, but could be valuable to guide the development of new treatment options. The aim of this meta-analysis is to shed light on the degree that excitatory, inhibitory and modulatory neurotransmission systems are affected in DoC.

Methods: In June 2023, we used MEDLINE/Scopus/Embase to search for resting-state MRI and PET studies, involving adults with a clinical diagnosis of DoC based on a validated behavioural scale. Two referees screened studies and extracted coordinates of whole-brain, voxel-based comparisons performed between DoC patients and controls. Coordinate-based meta-analysis was performed via GingerALE. The resulting statistical map of voxel-wise likelihoods was compared to 33 neurotransmission templates, using the neuromaps toolbox. Spatial correlations were deemed significant at p<0.05 Bonferroni-corrected for multiple comparisons.

Results: Based on 39 studies in 1156 individuals, we computed a spatial map of likelihood of brain alterations in DoC. We identified subcortical (thalamus and caudatum) and cortical (angular gyrus, precuneus/posterior cingulate and medial frontal/orbitofrontal cortex) regions as highly likely to be structurally, functionally and/or metabolically affected in DoC. The likelihood of brain alterations in DoC was significantly and reliably associated with the distribution of glutamatergic mGLU5 receptors, GABA-A receptors, mu-opioid receptors, CB1 cannabinoid receptors, serotoninergic 1B receptors and noradrenergic transporters.

Conclusions: Alterations in different neurotransmission systems, with a variable involvement of specific pre- and post-synaptic elements in excitatory, inhibitory and modulatory systems, are associated with brain alterations in DoC. These exploratory findings should be confirmed by ad-hoc molecular imaging studies in DoC.

The recovery trajectory of patients with disorders of consciousness following traumatic brain injury (TBI) often involves a confusional state marked by varying intensity and duration, characterized by amnesia, disorientation, impaired arousal, and disturbances in attention. Despite its frequent occurrence, the neurophysiological mechanisms underlying recovery from this fluctuating state remain poorly understood. A common feature across pathological, pharmacological, and physiological alterations of consciousness is a slowing of periodic and aperiodic EEG activity.

We recruited patients with subacute TBI in a rehabilitation facility, recording EEG at admission (T0) from patients with post-traumatic confusional state (PTCS, N=20) or who already emerged from it (controls, N=13). We then longitudinally monitored the PTCS cohort with the Confusion Assessment Protocol and recorded another EEG (T1) upon emergence from confusion or at discharge.The 6-month outcome was assessed with the Glasgow Outcome Scale-Extended (GOSE) and Montreal Cognitive Assessment (MoCA).

Patients who emerged from confusion exhibited EEG recovery, renormalizing the spectral profile of both aperiodic and periodic activity (spectral exponent, spectral offset and peak frequency) to levels comparable to TBI controls. However, they showed residual abnormalities, including elevated broad-band amplitude extending into infra-slow frequencies, as indicated by the spectral offset. At T1, aperiodic EEG activity, particularly the spectral exponent, but not periodic activity, predicted cognitive and functional outcomes (MoCA and GOSE).

Overall, our findings reveal that PTCS is underpinned by broad-band slowing of both periodic and aperiodic neurophysiological activity, with long-term clinical outcome linked to aperiodic activity. These findings unveil the neurophysiological mechanisms of recovery from confusional state.

Introduction: Transcutaneous auricular vagus nerve stimulation (taVNS) is a promising, non-invasive neuromodulatory approach for disorders of consciousness (DoC), potentially restoring circuits critical for consciousness. By targeting afferent vagal fibers, taVNS may modulate scalp-level electroencephalographic (EEG) activity and influence cardiac parasympathetic control. Its specific bioelectrical signatures remain unknown but may underlie improved cognitive and autonomic regulation in patients with DoC.

Methods: This double-blind randomized controlled trial involved 44 DoC patients in the acute setting. Patients randomly received either bilateral active taVNS (30s alternating active/rest episodes; 3mA; 200-300μs width, 25Hz) or sham stimulation for 45 minutes daily over five days. Bioelectrical signals from high-density EEG and electrocardiogram (ECG) were collected at baseline and at the end of the intervention before, during and after the taVNS stimulations. We analyzed the signals’ metrics and compared their characteristics across conditions (active vs. sham) to determine taVNS’ safety profile and biomarkers of responsiveness.

Results: ECG analyses showed that bilateral taVNS was safe, with no significant group-level difference in mean heart rate (HR) between taVNS ON and resting periods (U=190; p=0.793). However, HR variability (HRV) increased during active taVNS (W=44; p=0.04) but remained unchanged during sham stimulations (W=68; p=0.293), supporting taVNS’ effect on parasympathetic fibers recruitment. Furthermore, we identified a treatment effect at the behavioral level (W=277.5; p=0.032). While planned EEG analyses did not reveal group-level differences, ongoing investigations aim to characterize the neurophysiological phenotype of responders.

Conclusion: These findings provide evidence supporting the safety and neuromodulatory potential of taVNS in acute altered states of consciousness.