Abstract
Detecting willful cognition in these patients is known to be challenging due to the patients’ motor disabilities and high vigilance fluctuations but also due to the lack of expertise and use of adequate tools to assess these patients in specific settings. This review will discuss the main disorders of consciousness after severe brain injury, how to assess consciousness and cognition in these patients, as well as the challenges and tools available to overcome these challenges and reach an accurate diagnosis.
Keywords
Introduction
Severe brain injuries leading to prolonged disorders of consciousness only account for a small proportion of all brain injuries with annual prevalence rates in the United States of around 300,000 cases (Giacino et al., 2020). However, patients with severe brain injuries come with a large cost (approximately US$4 billion per year in direct medical costs) due to their need for long-term care (CDC, 2019). It is commonly believed that severe brain injuries result in poor outcomes, but recent research has shown this to be untrue. Around 65% of patients with severe brain injuries who cannot respond to commands at the time of discharge from inpatient rehabilitation can function independently (for mobility and self-care) after 10 years post-injury (Hammond et al., 2019; Wilkins et al., 2019). In this context, to improve care for these patients and facilitate scientific research, the American Academy of Neurology (AAN) and the European Academy of Neurology (EAN) have published the first guidelines to help clinicians assess and care for patients with acute and prolonged (> 28 days) disorders of consciousness (DoC) (Giacino et al., 2018; Kondziella et al., 2020). Indeed, patients with DoC require specialized care and assessment. Detecting willful cognition in these patients is known to be challenging due to the patients’ motor disabilities and high vigilance fluctuations but also due to the lack of expertise and use of adequate tools to assess these patients in specific settings (Kondziella et al., 2020). This review will discuss the main disorders of consciousness after severe brain injury, how to assess consciousness and cognition in these patients, as well as the challenges and tools available to overcome these challenges and reach an accurate diagnosis. This review will nevertheless focus on behavioral techniques since other papers in this special issue address the use of neuroimaging and electrophysiology to detect conscious brain activity and covert awareness in patients with DoCs.
Disorders of consciousness
Coma
Following a severe brain injury, patients may remain unconscious for an extended period, referred to as a coma. Coma is a pathological state characterized by severe and prolonged dysfunction of vigilance and consciousness, resulting from either a brainstem lesion (involving the reticular activating system) or a global brain dysfunction caused by diffuse axonal injury. The key features of coma are the absence of eye-opening as well as the absence of oriented eye movements (i.e., visual fixation/tracking) and voluntary motor/verbal responses. It must last for at least an hour to distinguish it from a transient state such as syncope or acute confusion. Prolonged coma is rare and typically lasts 2 to 4 weeks before evolving into a vegetative or minimally conscious state (Posner et al., 2019).
Vegetative state
Patients in a vegetative state (VS) exhibit reflexive behaviors, such as spontaneous or stimulus-induced eye-opening (without sustained visual fixation) and preserved autonomic functions like breathing, cardiovascular regulation, and thermoregulation, but they lack consciousness (Table 1) (MSTF, 1994). They therefore start to show signs of arousal but in the absence of signs of consciousness of their environment (i.e., willful/oriented behaviors). Typically, VS results from an injury involving the white matter or bilateral lesions of the thalamus (i.e., intralaminar nuclei) (Monti & Sannita, 2016). Historically, the term “persistent VS” has been used for patients who have been in this state for more than a month, while “permanent VS” has been used for patients who have been in this state for more than 6 months after a non-traumatic brain injury (e.g., stroke or anoxia) or more than a year after a traumatic brain injury (MSTF, 1994). However, recent evidence suggests that some patients may recover even after a year, and the AAN recommends replacing “permanent VS” with “chronic VS,” with the duration specified (Giacino et al., 2018). In Europe, this clinical syndrome was originally (in the 40s) referred to as “apallic syndrome” (Laureys et al., 2010). However, this terminology has faced significant criticism, as contemporary research indicates that these patients are not truly “a-pallic” or lacking a cortex or pallium; instead, they possess isolated, albeit disconnected, islands of residual primary cortical functioning. The term “Unresponsive Wakefulness Syndrome” (UWS) has been suggested as a more neutral and descriptive term instead of both “apallic syndrome” and “vegetative state” by the European Task Force on Disorders of Consciousness (Laureys et al., 2010). Finally, they may also mimic emotional behaviors such as smiling, crying, grimacing or moaning but these should always be observed out of context (MSTF, 1994). All of these features may puzzle the family and complicate the work of the medical staff, who may be inclined to experience burnout. Therefore, providing accurate information and psychological support is essential to help families and medical staff cope with this challenging situation (Covelli et al., 2016).
Comparison of the behavioral features of coma, VS, MCS–, MCS+, EMCS and Covert Awareness (CA)
Comparison of the behavioral features of coma, VS, MCS–, MCS+, EMCS and Covert Awareness (CA)
The minimally conscious state (MCS) is a more recently identified condition than coma or the vegetative state. It is characterized by inconsistent but reproducible behavioral signs of consciousness (Table 1) (Giacino et al., 2002). This state marks the return of top-down cognitive processing such as attention, working memory and language processing as reflected by visual tracking, oriented movements and reproducible command-following (Martens et al., 2019). Resistance to eye opening and localization to sound has also recently been associated with consciousness but are still not officially recognized as such (van Ommen et al., 2018; Carrière et al., 2020). Visual fixation and tracking are usually the first signs of consciousness but can be difficult to detect and require the use of sensitive diagnostic tools (such as the use of a mirror) (Overbeek et al., 2018). Signs of consciousness in MCS patients can be inconsistent due to high vigilance fluctuations, but they must be replicated within a given examination to meet diagnostic criteria. Conducting serial examinations (
MCS has recently been subdivided into two clinical entities based on the presence (MCS+) or absence (MCS-) of receptive and expressive language (i.e., command-following, intelligible verbalization, and/or intentional communication) (Thibaut et al., 2020). The clinical subcategorization of MCS is supported by metabolic differences in areas that are associated with both consciousness (i.e., lower metabolism in the precuneus and thalamus in MCS-) (Aubinet et al., 2018a) and language (i.e., lower metabolism in the left middle temporal cortex and lower connectivity between left angular gyrus and left prefrontal cortex in MCS-) (Aubinet et al., 2018b; Aubinet et al., 2020). Several studies recently showed that, at discharge from acute rehabilitation, patients in MCS+ recovered consciousness the most and had the least disability (Aubinet et al., 2018a; Giacino et al., 2020). This also confirms that both level of consciousness and time since injury are important when considering prognosis in these patients (Estraneo et al., 2020).
Emergence from MCS (EMCS) is defined by the re-emergence of reliable and consistent functional communication and/or functional object use. Studies show that a majority of patients with severe brain injury recover functional communication within eight weeks post-injury (Martens et al., 2020). There also have been recent findings showing that consistent response to command is associated with emergence from MCS and should be recognized as such when updating the diagnostic criteria for MCS (Golden et al., 2022). When emerged from MCS, most patients evolve towards an acute confusional state characterized by cognitive impairment (98%), disorientation (93%), and agitation (69%) (Bodien et al., 2020).
Covert awareness
For the past decade, clinicians have been encountering a new type of patient showing covert awareness. These patients are unable to display any physical signs of consciousness but can demonstrate high cognitive processing since they are able to respond mentally to active neuroimaging or electrophysiological tests (see Table 1) (Schiff, 2015; Owen, 2019). These patients might be misdiagnosed as being in a Locked-in syndrome (LIS) (Schnakers & Laureys, 2018) which is caused by a localized ventropontine lesion resulting in paralysis of all four limbs without affecting consciousness or cognition. However, recent studies suggest that this clinical entity may be caused by impaired connectivity between the thalamus and primary motor cortex, which interferes with the execution of voluntary motor actions (Fernández-Espejo et al., 2015). Two meta-analyses revealed that this clinical entity is uncommon but not rare in patients with a vegetative state (14–17%), but more frequent in those with traumatic brain injury (Kondziella et al., 2016; Schnakers et al., 2020). Future studies will need to be conducted across multiple centers to gather enough data to establish a complete profile of this clinical condition. Moreover, even if this phenomenon has been known for more than a decade, there has been no agreement regarding the taxonomy to use for these patients who are able to respond to active neuroimaging or electrophysiological paradigms. A recent systematic review found 25 different names given to this entity across the literature (Schnakers et al., 2022). The five following names were the ones the most frequently used: covert awareness, cognitive motor dissociation, functional locked-in and non-behavioral MCS (MCS*). Future studies will have to help in reaching a standard taxonomy, which will be key to achieve a successful clinical translation that is crucially needed.
Previous research has demonstrated that some patients with covert cognition have the ability to communicate. Therefore, recent studies have been exploring the potential benefits of Brain Computer Interfaces (BCI) for severely brain-injured patients. These interfaces utilize neuroimaging or electrophysiological signals to enable communication for patients who are otherwise unable to express themselves (Gibson et al., 2016). However, BCI paradigms are currently hard to implement. The tasks used in BCI communication are complex and some patients may not be able to respond, even if they are conscious. To implement augmentative communication techniques successfully, future studies will need to gain a better understanding of the residual cognitive functioning of these patients (Schnakers et al., 2015a).
Assessment and diagnosis
The challenges
A series of studies have found that around 40% of patients believed to be in VS are misdiagnosed and are in fact conscious (Schnakers & Laureys, 2018). Diagnostic accuracy for differentiating MCS from VS can be impacted by various factors, including biases from the examiner, patient-related factors, and environmental factors. Standardized rating scales, such as the Coma Recovery Scale-Revised (see section below), can help reduce examiner error in diagnosing patients with MCS or VS (Wade, 2018). However, it is important for examiners to follow specific administration and scoring guidelines to ensure diagnostic accuracy. Serial assessments, meaning multiple assessments conducted over a short period of time, have also been recommended by both the AAN and EAN to reduce misdiagnosis (Giacino et al., 2018; Kondziella et al., 2020). Recent research has suggested that performing at least five assessments within a two-week period may reduce misdiagnosis compared to a single assessment. These assessments can be performed by one or several assessors as long as these assessors are trained or have enough experience in using the target behavioral scale in DoC population (Giacino et al., 2018). Additionally, using relevant stimuli, such as a mirror to detect visual pursuit or familiar objects, during assessments may also improve diagnostic accuracy (Heine et al., 2017; Sun et al., 2018). Actually, involving the family in the assessment should be considered as it might lead to an increased responsiveness (Schnakers & Laureys, 2018).
Fluctuations in arousal level, fatigue, seizures, acute infections with fever (e.g., urinary tract infection or aspiration pneumonia), pain (e.g., due to decubitus, fractures, heterotopic ossification), (cortical) sensory deficits (leading to visual and hearing impairments), sympathetic storming and motor impairment (e.g., generalized hypotonus, spasticity or paralysis) also decrease the probability of observing signs of consciousness (Schnakers & Laureys, 2018). Optimizing arousal is crucial for accurate consciousness assessment in patients with disorders of consciousness, such as those in VS or MCS. Fluctuations in arousal level can make it difficult to assess the patient’s level of consciousness, as they may have brief periods of wakefulness and responsiveness that can be missed during an assessment. The AAN and EAN guidelines emphasize the importance of optimizing arousal in patients with disorders of consciousness, through measures such as reducing sedation, avoiding medications that may depress consciousness, and providing sensory stimulation to increase wakefulness (Giacino et al., 2018; Kondziella et al., 2020). Recent studies have highlighted the impact of circadian rhythms on arousal and the potential use of body temperature to identify the optimal time to assess patients. For example, patients with disorders of consciousness often have disrupted sleep-wake cycles, which can affect their level of arousal and responsiveness during assessments. By monitoring their body temperature over time, it may be possible to identify their circadian rhythm and determine the best time to conduct assessments when their level of arousal is highest. However, further research is needed to determine the utility of this approach in clinical practice (Bareham et al., 2019; Blume et al., 2017). Finally, the environment in which the patient is evaluated may bias assessment findings. Restricted range of movement stemming from restraints and immobilization techniques, poor positioning, under-suctioning, no adequate rest prior assessment (care/therapy just before) and excessive ambient noise/heat/light can all decrease or distort voluntary behavioral responses (Schnakers & Laureys, 2018).
Behavioral scales
Given the frequent occurrence of behavioral fluctuations in this particular population, it is crucial to conduct repeated evaluations over time. These evaluations should utilize measures that are sufficiently sensitive to detect subtle yet clinically significant changes in neurobehavioral responsiveness. Traditional assessment methods conducted at the bedside and neurosurgical rating scales like the Glasgow Coma Scale (GCS) (Reith et al., 2016) have limited effectiveness in monitoring the progress of patients experiencing prolonged DoC. Such scales primarily identify noticeable alterations in behavior and do not possess the ability to differentiate between random or reflexive behaviors and those that are intentional. In the acute stage of severe brain injury, the Full Outline of UnResponsiveness score (FOUR score) has proven to be more sensitive than the GCS in detecting not only signs of consciousness but also various levels of brainstem dysfunction (Anestis et al., 2020). The EAN has hence recommended the utilization of the FOUR score to evaluate patients with severe brain injuries in an acute setting (Table 2) (Kondziella et al., 2020).
Responses assessed by behavioral scales commonly used and/or recommended to be used for DoCs
Rem: N = Noxious, A = Auditory, T = Tactile, O = Olfactory, G = Gustatory.
Responses assessed by behavioral scales commonly used and/or recommended to be used for DoCs
Rem:
The neurobehavioral assessment measures specifically designed for patients with prolonged DoC include the Coma Recovery Scale-Revised (CRS-R), the Western Neurosensory Stimulation Profile (WNNSP), the Western Head Injury Matrix (WHIM), the Sensory Modality and Rehabilitation Technique (SMART), the Sensory Stimulation Assessment Measure (SSAM), and the Disorders of Consciousness Scale (DOCS) (Table 2) (Seel et al., 2010). While the specific items covered may differ among these measures, they all assess behavioral responses to various prompts related to auditory, visual, motor, and communication stimuli. All of these instruments have demonstrated sufficient reliability and validity. However, there is notable variation in their psychometric properties and clinical usefulness.
Out of the mentioned measures, the Coma Recovery Scale-Revised (CRS-R) is unique in its direct incorporation of the established diagnostic criteria for VS and MCS within its administration and scoring process. In 2010, the American Congress of Rehabilitation Medicine conducted an evidence-based review of neurobehavioral rating scales specifically designed for patients with DoC (Seel et al., 2010). Six were recommended for clinical practice: the CRS-R, the WNNSP, the WHIM, the SMART, the SSAM, and the DOCS. The CRS-R received the strongest recommendation with “minor reservations” due to its excellent performance across various psychometric quality indicators. The EAN has also strongly recommended its use to assess DoCs (Kondziella et al., 2020). Additionally, the CRS-R is listed as one of the Traumatic Brain Injury (TBI) Common Data Elements (CDE) recommended by the US National Institute of Neurological Disorders and Stroke (NINDS) and is the preferred method for monitoring consciousness recovery in TBI research. Lastly, the scale has been translated into over 10 languages and is currently utilized worldwide.
The CRS-R scale comprises 23 items organized into six subscales that cover auditory, visual, motor, verbal/oromotor, communication, and arousal functions. Scoring follows standardized guidelines and is based on the presence or absence of operationally-defined behavioral responses to specific sensory stimuli. Psychometric studies have demonstrated that the CRS-R meets rigorous criteria for measurement and evaluation tools used in interdisciplinary rehabilitation settings. Trained examiners can administer the CRS-R reliably, and scores tend to remain relatively stable across repeated assessments. The CRS-R has also been adapted for use in the pediatric population. The CRS-Pediatrics (CRS-P) is suitable for children as young as 12 months of age (Slomine et al., 2019).
In subacute settings, the CRS-R has exhibited superior diagnostic accuracy compared to other consciousness scales like the GCS or the FOUR (Seel et al., 2010). Recent research findings indicate that higher CRS-R scores upon admission to inpatient rehabilitation facilities can help differentiate patients who will have better outcomes at discharge. As mentioned previously, this confirms that both level of consciousness (as reflected by CRS-R scores) and time since injury are important when considering prognosis in these patients (Estraneo et al., 2020). This information is valuable for rehabilitation planning and effective communication with patients and their caregivers (Portaccio et al., 2018a; Portaccio et al., 2018b). A study by Wannez and colleagues demonstrated that focusing on the five most commonly observed behaviors in the CRS-R assessment (i.e., fixation, visual pursuit, reproducible movement to command, automatic motor response, and localization to noxious stimulation) successfully identified 99% of patients in MCS (Wannez et al., 2018). Based on these findings, the Simplified Evaluation of CONsciousness Disorders (SECONDs) has recently been developed for allowing shorter but sensitive assessments and might be an ideal alternative in subacute settings where time is limited (Aubinet et al., 2021a; Sanz et al., 2021).
Individualized Quantitative Behavioral Assessment (IQBA)
When clinicians care for patients with MCS, they often face difficulties in interpreting the significance of the patients’ behavioral responses, which may be infrequent or ambiguous due to the patients’ sensory, motor, and arousal deficits. To address these issues, the Individualized Quantitative Behavioral Assessment (IQBA) was developed to provide tailored assessment procedures and control for bias (Whyte et al., 1999). IQBA involves defining a target behavior and recording its frequency in response to a command, an incompatible command, and during rest. It is then possible to determine whether the behavior occurs more often in one condition than the others, helping to differentiate between random movements and purposeful responses to commands. Day and colleagues demonstrated that IQBA can provide consistent responses to commands and detect consciousness earlier than the CRS-R in some patients. Therefore, IQBA approaches could be used in conjunction with the CRS-R to improve diagnostic accuracy (Day et al., 2018).
Pain assessment
Detecting whether a patient with DoC is experiencing pain is important for clinicians and families. However, self-reporting is not possible in these patients due to their inability to communicate. Currently, the Nociception Coma Scale-Revised (NCS-R) is the primary tool used to evaluate nociception and pain in patients with DoC. The term “nociception” was selected for two reasons: firstly, the NCS-R aims to evaluate patients in both VS and MCS, and thus assesses responses related to both low-level brain processing associated with nociception (as observed in most patients in VS), as well as high-level brain processing associated with pain (as seen in patients in MCS). Secondly, since pain is a subjective experience, it is challenging to use the term “pain” when self-report is not possible (Schnakers & Zasler, 2015b).
The Nociception Coma Scale (NCS) and its revised version (NCS-R) were developed using pre-existing pain scales validated for non-communicative patients with dementia and newborns (Schnakers et al., 2010; Chatelle et al., 2012). The NCS-R includes three subscales that assess motor and verbal functions, as well as facial expression, making it suitable for assessing nociception and pain in patients with DoC. Compared to other pain scales, the NCS-R has a broader score range and is more sensitive to clinical diagnosis of DoC, indicating its relevance for evaluating pain in these patients. Recently, a larger study confirmed the relationship between this scale and the level of consciousness (Chatelle et al., 2018).
Based on this revised version (score range: 0–9), Schnakers and coworkers defined a cut-off score of above 4 as a potential clinical threshold for detecting pain in DoC patients (sensitivity of 73%, specificity of 97% and accuracy of 85%) (Chatelle et al., 2012). Using this threshold, Chatelle and coworkers assessed the clinical usefulness of the NCS-R and showed, in DoC patients with potential painful conditions (e.g., due to fractures, decubitus ulcers, or spasticity), decreased total scores under analgesic treatment without a decrease of level of consciousness (Chatelle et al., 2016). In a recent neuroimaging study using a 18-Fluoro-deoxyglucose PET scan, the same investigators have found a significant correlation between NCS-R total scores, its pain threshold and brain metabolism in the pain network and more particularly with the ACC which is associated with the emotional processing of pain (Bonin et al., 2020). Those results suggest that the NCS-R is related to pain processing and constitutes an appropriate behavioral tool to assess, monitor and treat pain in non-communicative patients with DoC.
The NCS and its revised version have been translated into eight different languages, including English, French, Italian, Flemish, Danish, Portuguese, Russian, and Thai. They are used in various settings, from intensive care (Bernard et al., 2019) to long-term facilities (Poulsen et al., 2019). A recent systematic review found that both scales are valid and useful instruments for assessing pain in DoC patients based on their psychometric properties (Vink et al., 2017). Furthermore, the International Multi-disciplinary Consensus Conference on Multimodality Monitoring, which included the Neurocritical Care Society, the European Society of Intensive Care Medicine, the Society for Critical Care Medicine, and the Latin America Brain Injury Consortium, recommended the NCS-R for pain assessment in this patient population (Riker et al., 2014).
The NCS has been the first scale developed for assessing nociceptive pain in patients with severe brain injury. However, Whyte and his colleagues have recently developed another measure of nociception specifically for these patients called the Brain Injury Nociception Assessment Measure (BINAM). Preliminary results suggest that the BINAM is a reliable and feasible tool to assess the intensity of nociception, independent of the level of consciousness, which is not the case with the NCS-R. More data is needed to further establish the psychometric properties of the new scale (Whyte et al., 2020).
Language assessment
The current behavioral scales used to evaluate cognition in individuals with DoC lack the ability to identify precise cognitive impairments. Schnakers and colleagues have demonstrated that assessing consciousness is challenging due to the simultaneous presence of cognitive deficits like difficulties in receptive language. This highlights the importance of creating novel tools and scales to accurately assess these impairments in these patients (Schnakers et al., 2015c).
A new cognitive assessment tool called the Cognitive Assessment by Visual Election (CAVE) has been developed specifically for patients with DoC (Murphy, 2018; Aubinet et al., 2018a). The CAVE consists of six subscales, each comprising ten items, which evaluate the recognition of real objects, numbers, written words, letters, pictures, and colors. During the test, the patient is instructed to focus on a specific target while ignoring any distractions. Since the test requires the ability to maintain visual fixation, it is designed for patients in MCS minus, MCS plus, and emerging from MCS. On average, the administration of the CAVE takes between 10 and 30 minutes. Preliminary findings have shown a high level of agreement among different raters, indicating strong inter-rater reliability, along with satisfactory internal consistency (Murphy, 2018). Additionally, CAVE scores appear to decrease as the CRS-R total score decreases, establishing a consistent behavioral and cognitive profile for each patient. Lastly, patients with higher CRS-R and CAVE scores exhibited less pronounced cerebral hypometabolism in the language network (Aubinet et al., 2018b). Finally, the Brief Evaluation of Receptive Aphasia (BERA) was created with the intention of assessing remaining language skills in a detailed and thorough manner. Its purpose is to differentiate between receptive phonological, semantic, and morphosyntactic abilities by observing the visual fixation on target stimuli amidst phonologically or semantically similar distractions. While this scale has been validated for language-specific evaluation among individuals with aphasia, its complete validation in patients with DoC is nevertheless still pending (Aubinet et al., 2021b).
Conclusion
Misdiagnosis can lead to inappropriate treatment and management decisions, including withdrawal of life-sustaining treatment in patients who may have the potential for recovery. Therefore, it is essential to employ a comprehensive approach to clinical assessment to minimize the risk of misdiagnosis and ensure that patients receive appropriate care. Collaborative efforts between clinicians, researchers, and families of patients with DoC may facilitate the development and implementation of improved assessment methods. Recent preliminary findings show promise in assessing language functions in DoC patients using new assessment tools such as the CAVE and the BERA. However, assessment tools still need to be developed to allow an assessment of specific cognitive components in patients with DoCs. Developing and validating tools that can assess more precisely language and cognition in DoC patients will be crucial for better tailoring rehabilitation programs and improving treatment outcomes for these patients.
Footnotes
Acknowledgments
The author would like to thank the Casa Colina Board of Directors and the Casa Colina Foundation for supporting the research. The foundation was not involved in writing the manuscript.
Conflicts of interest
The author declares no conflict of interest.
Funding
None to report.
Ethical approval
Not applicable.
Informed consent
Not applicable.
