A New Way To Test For Signs Of Consciousness In Unresponsive, Brain-Injured Patients

Patients who are brain-injured and unresponsive may appear unconscious, but a study published Wednesday in the New England Journal of Medicine repurposed a widely-used technology to demonstrate that the brains of some of these patients are still active.

The researchers used electroencephalography or EEG to look for signs of brain activity in a group of brain-injured patients, finding that 15% of those studied had residual activity despite being unable to speak or move.

EEG is already used to diagnose epilepsy and other brain disorders, but this study shows that EEG recordings can be used to detect what some researchers call "preserved consciousness" in some unresponsive patients with a severe brain injury. This method might make it easier for doctors to predict whether a brain-injured patient will wake up from a coma and might help inform decisions related to withdrawal of life-support.

"I think this is a landmark study with the potential to impact clinical practice." says Dr. Brian Edlow, director of the Laboratory for NeuroImaging of Coma and Consciousness at Massachusetts General Hospital, who was not involved with this research. "Because EEG can be performed at the bedside, I think it has the potential to be generalized in ICUs around the world."

Dr. Jan Claassen, head of neurocritical care and medical director of Columbia University's neurological intensive care unit, is the lead author. He became interested in understanding consciousness after brain injury when he discovered a different study that used functional magnetic resonance imaging (fMRI) to detect brain activity in brain-injured, apparently unconscious patients in response to verbal commands.

This prompted him to start his own investigation, this time using EEG, a cheaper and more widely available technique.

"I got interested in consciousness very early on in my medical career," says Claassen. "[From previous studies], it became clear that some patients that appear to be unconscious because they can't move or talk, respond in a similar way to patients that can express that they understand [verbal] commands."

Claassen and the other researchers used EEG to measure the brain activity of 10 healthy volunteers and 104 patients with an acute brain injury who were unable to respond to spoken commands. They used headphones to play a list of simple verbal instructions for each patient, including cues to open and close, and to stop opening and closing, each hand.

They analyzed the EEG readings using an algorithm and found that 15% (16 of 104) of the brain-injured patients showed brain activity in response to verbal commands. And a year after injury, 44% (7 of 16) of patients who had shown brain activity were able to be at home without any assistance for up to eight hours. Only 14% of those who did not show brain activity were able to achieve that degree of independence.

Davinia Fernández-Espejo, a senior lecturer in the school of psychology at the University of Birmingham in the United Kingdom, who wasn't involved in the study, calls it "elegant" and "very important."

Previous studies have shown that approximately 15% of patients who have been brain-injured and unresponsive for months or years are actually aware, and their brains respond to external commands, she explains. But this study found brain activity in 15% of patients very soon after injury, within about four days.

"This is crucial because it's in these early days when important medical decisions related to withdrawal of life support take place. The information that we can obtain about individual patients from neuroimaging tests like the one reported here has great potential to inform these decisions," said Fernández-Espejo in an e-mail statement to NPR.

It's unclear whether unresponsive patients who showed brain activity could actually understand the verbal commands, or whether they were simply responding to the investigator's voice. "That is a very important and fascinating question, but we really don't know," Claassen explains. "We know that the brain reacted differently to the command 'open your hand' and 'stop opening your hand', [but we don't know] how much they understood."

Brain activity in these unresponsive patients might be used to predict whether they will regain consciousness. Half of the 16 patients who showed brain activity recovered enough that they were able to follow the same verbal commands before they were discharged from the hospital to continue rehabilitative treatment elsewhere. Only about a quarter of the patients who did not show brain activity were able to follow the verbal commands before they left the hospital.

Claassen thinks EEG can offer a much better alternative for monitoring brain activity in unresponsive brain-injured patients, compared to the current choices.

"It's a lot of logistical effort to move patients with a brain injury to an MRI scanner, and we have to weigh the safety of the patient against the information that we will get. The advantage of EEG is that it can be repeated multiple times and it's cheaper than an MRI scan," says Claassen.

Although the results are promising, Fernández-Espejo cautions that there are issues to address before the use of EEG becomes a widespread practice for detecting brain activity in apparently unconscious patients. "These results ... need to be validated in larger samples, and studied in the context of other existing prognostic tools available before their potential for clinical use can be evaluated appropriately," she said via e-mail.

Data analysis of the EEG data is also a problem. Any ICU would have the ability to use EEG to gather data on patients' brain waves, but the data analysis requires expertise and qualified personnel, Edlow explains.

"One can imagine a scenario whereby ICUs acquire the data and send it to specialized centers that have the capability to process and analyze it," he says. "I think this would be a way forward for ICUs to implement this type of technique."

Claassen and his group plan to continue following the patients from the study in the long-term. "My hope is that we get better at predicting who recovers from a coma and who doesn't," he says.

Luisa Torres is a AAAS Mass Media Fellow on NPR's science desk. She's on Twitter at @luisatorresduq.

Copyright 2019 NPR. To see more, visit https://www.npr.org.