ARTICLE IN BRIEF:
Using fMRI in people with disorders of consciousness and healthy controls, investigators were able to identify fluctuating patterns of activity in certain brain regional networks. The patterns offer possible biomarkers for consciousness.
A distinct pattern of coordinated activity across widespread brain areas may serve as a “fingerprint” of consciousness, both in patients with disorders of consciousness and healthy controls, a large multicenter study has found.
By contrast, when coordination is low and seen primarily between nearby structurally connected areas, but not beyond, it appears to be is a marker of unconsciousness, according to the study published February 6 in Science Advances.
The high- and low-connectivity patterns were merely the most extreme of four patterns in all that the study discerned, including two in-between patterns that were less clearly associated with either consciousness or unconsciousness.
Perhaps the most surprising finding of the paper was that all 159 subjects—including those who were minimally conscious, unresponsive, and healthy controls—were continuously shifting between the four brain states. Those showing clinical signs of awareness simply spent much more time in the widely connected state and had more chances to visit the other patterns, while unconscious individuals spent most of their time in the low-connectivity pattern from which they departed only briefly.
Study Design, Findings
Unlike most previous attempts to discern brain activations indicative of consciousness in people with disorders of consciousness, the functional MRI study examined their brain activity in a resting state, rather than hoping to catch them paying attention to the sounds of family members speaking or following mental-imagery instructions. As such, the resting-state findings may represent a less stringent, direct marker of awareness, permitting identification of patients who are at least somewhat alert, even if not able to follow the demands of mental-imagery tasks. In other words, the study mostly quantified the brain's capacity for consciousness rather than reporting on the contents of conscious experience.
The tests were carried out in four separate laboratories around the world—France, Belgium, Canada, and the United States—and the patterns were consistent throughout.
“The findings are so robust that different teams at different medical centers, using different tools, still had almost completely similar results,” said a coauthor of the paper, Nicholas D. Schiff, MD, professor of neurology and neuroscience at Weill Cornell Medical College in New York. “That's never before been shown for any measure of consciousness. This is the first demonstration of cross-platform reliability.”
Beyond the clinical implications for developing objective, quantifiable standards for assessing awareness in patients with disorders of consciousness, the researchers and other neurologists familiar with the paper emphasized, the findings offer something more fundamental: empirical evidence pointing to how consciousness emerges from three pounds of brain matter.
“What you need in any object of scientific study is something you can measure that is robustly related to the phenomenon,” Dr. Schiff said. “This gives us quantitative information about brain activity that is specifically associated with consciousness. That's very important to science. It's something we can now explore further.”
The study built on years of research, including a 2015 paper in Proceedings of the National Academy of Sciences showing that wakefulness in monkeys is characterized by both positive and negative correlations among far-flung brain regions, whereas unconsciousness during anesthesia is marked by connectivity of only anatomically connected regions.
“I had done a resting-state fMRI study of patients, but with the signals averaged over 10 minutes, as part of my doctoral research at the University of Liège in Belgium,” said the first author of the new paper, Athena Demertzi, PhD, now a research associate at the Belgian Funds for Scientific Research at the University of Liège in Belgium. “During my postdoctoral studies at the ICM Brain and Spine Institute, in Paris, France, we went a step further. My colleagues there had just published the study in monkeys. They described how the signal changed across time. With my resting-state paradigm, and their expertise in the dynamic patterns, we said, let's figure out how to combine these approaches in patients.”
The study included 21 healthy individuals and 63 patients in Liège, 15 healthy individuals and 22 patients in Paris, 11 healthy individuals and 16 patients in New York, and 11 patients in London, Ontario. Patients were diagnosed as either being in a vegetative state—also known as unresponsive wakefulness syndrome (UWS)—or in a minimally conscious state (MCS), based on repetitive, standardized behavioral assessments. Patients and healthy individuals alike were scanned both under anesthesia with propofol and while sedation-free.
Their resting state was interrogated with fMRI BOLD signals that explored the coordination between 42 non-overlapping regions of interest representing six brain networks implicated in functional and cognitive processes.
Dr. Demertzi and colleagues found four distinguishable patterns of coordination among the regions of interest.
“A pattern of high complexity (pattern 1), including positive and negative values of long-distance coordination, was more prevalent in healthy participants and in patients in MCS as compared to patients in UWS,” the paper stated. “The rate of this pattern also increased when moving from patients in UWS, to patients in MCS, and to healthy control individuals. In sharp contrast, a pattern of low inter-areal coordination (pattern 4) was more likely to occur in unresponsive patients compared to patients in MCS. This pattern also presented a decreasing probability rate from patients in UWS, to patients in MCS, and to healthy control individuals.”
The robustness of the four patterns was tested and confirmed by separate analysis of the results from each of the four scanning sites.
The investigators also analyzed the results in a group of the UWS patients who underwent a separate fMRI test in which they were asked to imagine themselves playing tennis or not during seven challenges, a method first described in 2006 by Dr. Adrian Owen, now of the University of Western Ontario.
Patients who showed they were capable of following the commands, despite evincing no behavioral evidence of consciousness, were more able to occupy the high-complexity pattern 1 during the resting state as compared with those UWS patients who did not execute the task. In contrast, these UWS patients who were not able to follow mental-imaging commands tended to maintain a low-complexity resting-state pattern, similar to that seen in anesthetized patients.
Researchers who specialize in the study and treatment of disorders of consciousness praised the study.
“Some people might view this paper as offering a new test of consciousness, but in my view that's not its main contribution,” said Joseph T. Giacino, PhD, director of rehabilitation neuropsychology and research associate at Spaulding Rehabilitation Hospital and associate professor in the department of physical medicine and rehabilitation at Harvard Medical School. “Really, it gives us a better idea as to how the brain generates consciousness, as guided by global workspace theory.”
The fact that both correlations and anti-correlations are seen in pattern 1, he said, suggests that the brain is selecting which areas to attend to, and which to ignore, on a dynamic basis.
“Everything can't be the center of attention, so how does the brain attend to one thing versus another, so you don't walk off a cliff as you're talking to somebody?” Dr. Giacino said. “What this paper is showing is that the healthy conscious brain is continuously tuning its dials to coordinate its signals.”
It's possible, he said, that the new measure could be the basis of an index for the probability of consciousness in patients with disorders of consciousness, similar to the perturbational complexity index described in a 2013 report in Science.
Brian L. Edlow, MD, director of the Laboratory for NeuroImaging of Coma and Consciousness at Massachusetts General Hospital, and assistant professor of neurology at Harvard Medical School, said the new study “significantly advances our understanding of the functional network connectivity properties that are necessary for human consciousness.
“It provides empiric evidence about the importance of integration of brain network nodes over long distances,” he added, “which supports the leading conceptual models of consciousness.”
While the study did not examine the subjective correlates of the four brain patterns it discovered, Dr. Edlow said, future investigations may do so.
“We have to recognize that this paper, while tremendously impactful for our field, has not demonstrated the clinical utility of these tests,” he said. “That clinical validation awaits further studies.”
Regarding such studies, Dr. Dermertzi said: “I am very interested to go a step further and give cognitive meaning to these patterns. We are now preparing an experiment where we plan to have a direct mapping between cognitive function and the brain configurations with the aim to ‘brain-read’ resting acquisitions and be able to quantify some basic cognitive components, which is a key clinical imperative.”
Drs. Dermertzi, Schiff, Edlow, and Giacino had no relevant disclosures.