ARTICLE IN BRIEF
A research team found that one in five patients who had been assessed to be in a vegetative state showed evidence on an EEG recording that they understood the commands to do a mental task.
A team of Canadian scientists has developed a bedside EEG task to assess whether a person in a vegetative state can carry out mental exercises, evidence of conscious awareness even though their bodies reveal no overt signs of responsiveness.
In the study, published in the Nov. 10 online edition of the Lancet, one in five patients in a vegetative state (VS) showed evidence on an EEG recording that they understood the commands — squeezing their hand and wiggling toes — over and over again during a 20-minute testing period.
Adrian M. Owen, PhD, the Canada Excellence Research Chair at the University of Western Ontario, Damian Cruse, PhD, a post-doctoral fellow at the University of Western Ontario, and their colleagues tested the cognitive abilities of 16 patients in a VS — observed in two inpatient European research units, the Addenbrooke's Hospital in Cambridge and University Hospital of Liège, Belgium — for a week between July 2010 and June 2011. The patients were between 29 and 45 years old and had been in a VS from three to 23 months.
Three of the patients consistently and reliably generated EEG patterns identical to the patterns observed when 12 healthy volunteers did the same mental task — imagining that they were squeezing their right fist or wiggling their right toe. In the three patients, the premotor cortex, which is involved with imagined movement and planning, was consistently activated during the moments they were asked to carry out the task.
The EEG wave activity — as a measurement of mental awareness — could be a useful tool for the clinical assessment of VS patients, said Dr. Cruse. The patterns of brain activity suggest that some of these patients have the intention or desire to move.
“There was nothing about these patients that would lead one to believe that this would happen,” said Dr. Cruse. “We found that they were responding to our commands and that meant that they understood what we were asking, paying attention to the complex task and able to follow through with the command. This suggests that they are aware of what is going on in their environment and would be able to communicate if they had the right tools.”
Ultimately, he said, “we may be able to use this type of task to begin asking them if they know where they are and what they know about their lives.”
The investigators said EEG testing, which is more affordable and accessible than fMRI, could be used more to assess VS patients at bedside or even in their homes, but there are technological challenges that need to be resolved first. For example, the scientists are now working with engineers to figure out how they can analyze the EEG recordings in real time so that they can have an ongoing exchange with patients or caregivers.
The investigators are also trying to develop the technology so that patients who show conscious awareness without any overt behavioral proof might someday be able to communicate their needs and thoughts through a computer interface, similar to the tools available for patients with complete paralysis.
Dr. Owen and his colleagues have previously shown that fMRI scans could also be used to assess whether patients in a PVS can follow commands. In a study published last year in the New England Journal of Medicine, the researchers reported that four of 24 patients were able to answer “yes” and “no” questions by pairing that response with an imagined behavior such as imagining they were walking around their house or playing tennis.
“It is not at all practical to use fMRI scans to assess whether patients have conscious awareness and can communicate,” said Dr. Cruse. “The idea of making this type of assessment at the bedside was compelling.”
“Our studies prove that simply looking at behavior does not tell you that patients are aware,” he said. He added that the absence of a response with this task does not mean that there is a lack of awareness, either. This was a one-time assessment and it is impossible to tell whether the patient has periods of awareness at other times.
The investigators will monitor the patients over the year to see whether there is prognostic value in the initial data. They can assess whether these patients have better outcomes over time compared with those patients who showed no evidence of conscious awareness.
“The research we are doing is proof of principle,” said Dr. Owen. “Functional MRI is a fantastic technology but it is not portable.”
“This is the first step,” he said. “This is a challenging clinical problem. We are hoping that EEG assessment can help contribute to the proper diagnosis of vegetative state.”
EXPERTS WEIGH IN
In an accompanying editorial in the Lancet, Morten Overgaard, PhD, of the department of communication and psychology at Aalborg University in Denmark and Rikke Overgaard, PhD, of the CNRU Centre of Functionally Integrative Neuroscience, said that the study “is surprising and challenging. These findings present good evidence that at least some patients in the vegetative state are conscious. However, the methods in all these experiments are indirect and investigate a factor other than consciousness alone.”
“So far, most researchers have interpreted these published results as suggesting that many patients in the vegetative state are wrongly diagnosed,” they wrote. “However, these studies have an even stronger consequence. The real underlying issue is that the levels of consciousness have little to do with consciousness—i.e., subjective experience. A more plausible interpretation is that vegetative and minimally conscious states distinguish between different levels of cognitive and communicative abilities, which is a different matter than subjective experience. A new classification system is necessary if the goal is to understand the cognitive functioning of patients in the vegetative or minimally conscious states.”
Nicholas D. Schiff, MD, professor of neurology and neuroscience at Weill Cornell Medical College, and his colleagues also conducted an EEG study to demonstrate awareness in patients with severe brain injury. Their study, published online in April in Clinical Neurophysiology, assessed EEG recordings during a mental imagery exercise in three patients with a range of brain injuries — from minimally conscious state (MCS) to locked-in state. They found that one patient in MCS and one in the locked-in state showed evidence that they understood the commands.
“We are now at a point where there are a lot of good published data from several different research groups,” said Dr. Schiff. “Given that several studies have now found such patients in apparent vegetative states with command-following abilities, the question here is whether this particular study is more than additive for these new approaches.”
“Lay people will invariably interpret this paper, as the others, as ready for clinical use, something none of these methods has been validated for to date,” he said. “The findings reported in this paper are not so convincing that this paradigm will perform well as a clinical test. The EEG analysis could only pick up 75 percent of the controls.
“As the paper clearly will lead to discussion of using these and other tools to separate the MCS from vegetative states, the results should not convey the impression that any of our available research tools are yet clinically validated.”
The great advantage of the EEG is that it is a physiological signal that is generated by neuronal activity as opposed to changes in blood flow or oxygenation level typical of functional neuroimaging techniques, Dr. Schiff continued. As a consequence, changes in EEG responses can often be related to measures obtained from single-unit, local field potentials and electrocorticography allowing the physiological plausibility of the signal found in an EEG measurement to be vetted against strong correlative data.
“In the present study the actual EEG responses are hidden and not examined,” Dr. Schiff said. “We can only rely on the performance of the classifier as the reporter of a positive result. This can bring false positive findings here as classification algorithms can be very sensitive and over fit noise in EEG data. Both showing the actual differences in the signals classified and applying a multiple comparisons statistical test to mitigate this kind of error would have strengthened the impact of the study.”
That said, he added that “we are now at a point in the work of this subfield where there are many studies and leads to follow to try and validate methods, and the implications are the same for other EEG-based assessment paradigms as they have been. The problem is that there is currently very little infrastructure to provide clinical follow-up for patients with severe brain injury and structure further clinical engagement.”
Want to hear about the EEG testing of -patients in a vegetative state? Damian Cruse, PhD, talked to -Neurology Today about what the investigators did and found. Look for the podcast and a video -demonstrating their experiment on www.neurotodayonline.com.