On communicating through the coma-like state

A study just published in the New England Journal of Medicine reports on how a subset of patients diagnosed as being in a coma-like state can be trained to show specific brain activity to answer yes / no questions despite seeming to be unconscious and unresponsive.

Many news reports seem to suggest that researchers have found a way of ‘reaching inside coma’ with a brain scanner to communicate with patients but the findings are much more modest, only 5 out of 54 patients could reliably produce specific brain activity on command and only one was tested who could answer simple yes / no questions in this way.

Despite this, the study is still incredibly impressive and it indicates that some patients who seem unconscious may have a much richer inner life than we assume and it may be possible to communicate with some of them by measuring their brain activity.

The researchers put people in brain scanners and, in one condition, asked them to imagine standing still on a tennis court while swinging an arm to “hit the ball” back and forth to an imagined instructor, and in the other, to imagine navigating the streets of a familiar city or to imagine walking from room to room in their home. These were chosen because they show distinct patterns of brain activity on a scan.

This was tested both on healthy people, for a comparison of how activity should normally look, and in brain injured patients in a coma-like state.

Only 5 of the 54 patients responded with distinct brain activity, similar to the type found in all the healthy comparison participants, but in this subset, it indicated that they were likely following simple verbal commands.

This has been established before, but one criticism of these past studies was this this could just be an automatic response to the words in the command. We know that the brains of unconscious people respond briefly but automatically to words, even the person is not aware of hearing them.

The brain activity for the ‘tennis’ and ‘walking’ commands was much longer and more sustained than we might expect from the normal automatic response to words, so this was unlikely, but you might still argue that these are automatic, non-conscious responses.

To rule this out in one patient, the researchers asked six yes/ no questions about simple personal details and instructed the patient to imagine tennis for yes and walking for no.

Crucially, during the questions, the researchers prompted the patient with just the word “answer?”, meaning any different reactions that showed up couldn’t be just an automatic response to the word itself which was always the same.

Out of these six simple questions, the patient ‘responded’ correctly to 5, suggesting that they were genuinely understanding, considering and making a conscious response. This was in a patient who had no external signs of consciousness.

The scans for a couple of the questions are in the image above (click for a bigger version). You can see how different the responses are, but also how serious the brain damage is.

Importantly, these correct answers do not necessarily mean that the patient was completely mentally fine but ‘trapped’ their body. One common test used on definitely conscious patients after brain damage asks lots of these yes / no type questions (like “Do cinemas show films?” / “Are bottles edible?”) to test understanding.

Some patients can be fully conscious but their language so damaged that they can’t answer these questions, others can manage the less complex ones (the easiest are usually simple personal details) but not others, for the same reason.

All of the patients in coma-like states were clearly very brain damaged, so it could be that even the one who could make conscious responses might not have full understanding. On the flip side, it could also mean that some of the other patients may have been conscious but could not understand the task, and so did not show up on this test. You can see it’s a tricky area.

However, the discovery that it is possible to communicate, even in a simple terms, with a patient previously though to be in a coma is huge news and this research is likely to lead to further work trying to detect which patients are conscious and to develop methods to communicate with them.

Link to study summary in NEJM
Link to good write-up from New Scientist.