Beyond alpha power: EEG spatial and spectral gradients robustly stratify disorders of consciousness

Abstract

Neurophysiological markers can overcome the limitations of behavioural assessments of Disorders of Consciousness (DoC). EEG alpha power emerged as a promising marker for DoC, although long-standing literature reported alpha power being sustained during anesthetic-induced unconsciousness, and reduced during dreaming and hallucinations. We hypothesized that EEG power suppression caused by severe anoxia could explain this conflict. Accordingly, we split DoC patients (n = 87) in postanoxic and non-postanoxic cohorts. Alpha power was suppressed only in severe postanoxia but failed to discriminate un/consciousness in other aetiologies. Furthermore, it did not generalize to an independent reference dataset (n = 65) of neurotypical, neurological, and anesthesia conditions. We then investigated EEG spatio-spectral gradients, reflecting anteriorization and slowing, as alternative markers. In non-postanoxic DoC, these features, combined in a bivariate model, reliably stratified patients and indexed consciousness, even in unresponsive patients identified as conscious by an independent neural marker (the Perturbational Complexity Index). Crucially, this model optimally generalized to the reference dataset. Overall, alpha power does not index consciousness; rather, its suppression entails diffuse cortical damage, in postanoxic patients. As an alternative, EEG spatio-spectral gradients, reflecting distinct pathophysiological mechanisms, jointly provide a robust, parsimonious, and generalizable marker of consciousness, whose clinical application may guide rehabilitation efforts.

Neural complexity and the spectral slope characterise auditory processing in wakefulness and sleep

“…Evidence of the prognostic and diagnostic value of quantification of AEPs and neural information content or the spectral slope for assessing consciousness levels or the integrity of neural functions in reduced consciousness conditions is growing (Aellen et al, 2023; Alnes et al, 2021; Bekinschtein et al, 2009; Benghanem et al, 2022; Colombo et al, 2019, 2023; Comanducci et al, 2020; Liu et al, 2022; Morlet & Fischer, 2014; Pfeiffer et al, 2018). Measures based on information content can differentiate levels of consciousness (Casali et al, 2013), whereas both measures based on information content, spectral slope and AEPs are promising as prognostic tools for patients with disorders of consciousness (Colombo et al, 2023; Fischer et al, 2004; Liu et al, 2022; Pfeiffer et al, 2018; Tzovara et al, 2013). Recent work has shown that in an acute post‐anoxic coma, the complexity of auditory‐evoked EEG responses and the spectral slope can be either higher or lower than that of healthy and conscious controls (Alnes et al, 2021), opposite to what one would have expected based on studies evaluating the complexity or spectral slope of spontaneous neural activity (Lendner et al, 2020; Medel et al, 2022; Miskovic et al, 2019; Schartner et al, 2015; Timmermann et al, 2019).…”

“…Evidence of the prognostic and diagnostic value of quantification of AEPs and neural information content or the spectral slope for assessing consciousness levels or the integrity of neural functions in reduced consciousness conditions is growing (Aellen et al, 2023;Alnes et al, 2021;Bekinschtein et al, 2009;Benghanem et al, 2022;Colombo et al, 2019Colombo et al, , 2023Comanducci et al, 2020;Liu et al, 2022;Morlet & Fischer, 2014;Pfeiffer et al, 2018). Measures based on information content can differentiate levels of consciousness (Casali et al, 2013), whereas both measures based on information content, spectral slope and AEPs are promising as prognostic tools for patients with disorders of consciousness (Colombo et al, 2023;Fischer et al, 2004;Liu et al, 2022;Pfeiffer et al, 2018;Tzovara et al, 2013).…”

“…The second school of quantitative measures of consciousness assesses instead the information content of neural activity in the absence of sensory input, either in spontaneous activity (Frohlich et al, 2022;Schartner et al, 2017) or following cortical perturbations, most commonly using transcranial magnetic stimulation (Casali et al, 2013;Sarasso et al, 2021). This line of work has shown that when consciousness diminishes, for instance during sleep, anaesthesia, or in disorders of consciousness, the complexity of neural activity is diminished (Andrillon et al, 2016;Mateos et al, 2018;Medel et al, 2022;Schartner et al, 2015Schartner et al, , 2017, whereas the spectral slope typically steepens (Colombo et al, 2019(Colombo et al, , 2023Lendner et al, 2020;Medel et al, 2022). To this line of research, our results add that the reduction of complexity observed in NREM sleep can be explained by the density of local slow wave activity.…”

“…Similar to neural complexity, the spectral slope of the EEG power spectrum is also mostly evaluated in spontaneous activity and is found to be sensitive to changes in consciousness levels, such as those arising from sleep (Höhn et al, 2022;Kozhemiako et al, 2022;Lendner et al, 2020;Miskovic et al, 2019) and anaesthesia (Colombo et al, 2019(Colombo et al, , 2023Lendner et al, 2020). The spectral slope is computed in either low (generally below 20 Hz), or high frequencies (generally between 20 and 45 Hz) (Alnes et al, 2021;Colombo et al, 2019;Gyurkovics et al, 2022;Höhn et al, 2022;Kozhemiako et al, 2022;Lendner et al, 2020).…”

Assessing mismatch negativity (MMN) and P3b within‐individual sensitivity — A comparison between the local–global paradigm and two specialized oddball sequences

“…The most obvious candidate for this purpose is restingstate EEG, whose visual inspection is recommended for the standard evaluation of patients with DoC (Kondziella et al, 2020). Despite its high specificity, however, restingstate EEG methodology might lack the desired sensitivity to stratify patients with DoC (Hofmeijer et al, 2015;Estraneo et al, 2016; but see also Colombo et al, 2023). TMS-EEG approaches can effectively compensate for this (Casarotto et al, 2016(Casarotto et al, , 2023, but access to this technique is still limited to a few research centers.…”