Background
Severe brain injuries following traumatic or non-traumatic etiologies can lead to Disorders of Consciousness (DoC), a class of neurological conditions characterized by pathological alterations of arousal and/or awareness. DoC patients are highly heterogeneous regarding type, extent, and location of the underlying structural and functional brain pathology as well as in terms of their clinical phenotype [
1]. The DoC range from coma (patients’ eyes closed) to unresponsive wakefulness syndrome (UWS; eyes open either passively or in response to stimulation) - also known as vegetative state (VS) - to minimally conscious state (MCS; divided into “MCS-“: visual pursuit oberservable without signs of command following, and “MCS+”: command following without effective communication) [
2,
3].
The diagnostic and prognostic tests available for DoC in the current clinical practice demonstrate an unacceptably high misdiagnosis rate of approximately 40% [
4]. Although diagnostic accuracy can be improved by the application of well-established standardized behavioral rating scales such as the Coma Recovery Scale-Revised (CRS-R) [
5], these scales still lack optimal sensitivity (as they require the preservation of patient’s sensory, motor and executive functions) [
6,
7], and specificity (due to the lack of a ground-truth about consciousness in many patients with DoC) [
8]. Hence, behavioural assessments may fail to detect consciousness in DoC patients due to injury-induced individual cognitive, sensory or motor deficits. Meta-analyses suggest that up to 15–20% of DoC patients are likely to exhibit covert signs of consciousness or cognitive-motor dissociation (CMD) [
9], while seeming unresponsive purely based on behavioral bedside examinations [
10,
11].
Current guidelines on the diagnosis of DoC recommend multimodal evaluations using a combination of standardized neurological examinations, (functional) neuroimaging, and neurophysiological tools in order to improve and individualize diagnostic and prognostic accuracy [
12]. In particular, using innovative functional neurodiagnostic methods, which do not require behavioral command following of the examined patient, covert markers of consciousness can be detected and, in turn, diagnostic accuracy can be increased [
13,
14]. Among these recommended methods, quantitative electroencephalography (qEEG), functional magnetic resonance imaging (fMRI), and EEG combined with transcranial magnetic stimulation (TMS-EEG) have shown promising results [
10,
12,
15,
16]. Combining evidence obtained from different neurodiagnostic multimodal techniques and analyzing them with modern machine learning-based methods can further increase diagnostic accuracy [
14,
17].
However, despite major progress in accurate diagnosis and outcome prediction for DoC patients over the last years [
10], as to date, reliable prognostic indices and read-outs to prognosticate a patient’s clinical evolution are still missing [
18]. Moreover, the etiology and pathophysiology of DoC are highly heterogeneous [
19], and most likely result from the combination of several factors, whose role and interplay still need to be clarified. Similarly, the precise neuronal mechanisms underpinning consciousness and its loss and recovery are still poorly understood, especially at the individual patient level. However, for ethical, therapeutic, and economic reasons, it is imperative to improve diagnostic accuracy and to predict outcome as early, reliably, and accurately as possible [
20‐
22]. Indeed, many medical decisions and further treatment or rehabilitation paths crucially rely on accurate diagnostic and prognostic results [
23,
24].
In addition, more reliable diagnosis and prognosis are of utmost relevance not only for patients’ management but also for the patients’ family members who very often become their informal caregivers. Taking care of DoC patients is a highly stressful experience for relatives and can represent a great emotional burden for them. Informal caregivers are exposed to substantial life changes and consequently often report low mental and physical health as well as high levels of distress [
25]. These burdens are related to the ambiguity of the situation they are facing, and especially to the uncertainty of its duration and outcome [
26,
27]. At the same time, they might need to act as surrogate decision-makers for their loved ones. Neuroimaging evidence of covert consciousness could confront them with ethical challenges [
28]. For these reasons, the development of personalized tools to improve diagnosis and prognosis of DoC as well as their comprehensive communication are of utter importance for both patients and their family members.
Aims
The overall goal of the multicenter European PerBrain project (full project title: “PerBrain: A Multimodal Approach to Personalized Tracking of Evolving State-Of-Consciousness in Brain-Injured Patients”) is to optimize the diagnosis and long-term prognosis of DoC patients following acute brain injury by combining clinical and multimodal tools. The project aims at developing a hierarchical diagnosis and prognosis procedure that allows a personalized description of the neurological status and the expected recovery potential at the single patient level.
The hierarchically applied methods include a wide range of the state-of-the-art neuroimaging-based (structural MRI and fMRI) and neurophysiological-based (qEEG and TMS-EEG) techniques as well as the investigation of brain-body interactions (recordings of nasal respiration during rest and during odor presentation). Further, well-established standardized clinical behavioral scales for the assessment of consciousness are administered (CRS-R, Glasgow Outcome Scale Extended GOS-E). Data from the different investigation modalities will then integrated by employing newly developed biostatistical analysis based on machine learning with the aid of information technology. In this way, a better understanding of the pathophysiological mechanisms of DoC is expected to be gained, consequentially enabling more tailored rehabilitation strategies and improving prognosis at single patient level.
In parallel to the optimal definition of the patients’ neurological status, PerBrain aims at investigating factors that impact well-being of informal caregivers of DoC patients and their treatment decision attitudes, specifically when confronted with results of multimodal technology-based tests, with the goal to develop strategies for the effective communication of technology-based results specifically tailored to caregivers’ needs.