Feb 04 2010

More on fMRIs and the Comatose

Dealing with patients in a coma is challenging in multiple levels. We are challenged to evaluate the degree of damage, or conversely the degree of neurological function that remains. We are challenged to give the family or caregiver an accurate prognosis. And we are challenged with dealing with the ethical and emotional issues that surround such cases. All of these challenges would be helped by improving our ability to accurately assess such patients – and fortunately we are making some progress in this area.

I have previously discussed research in which a woman in an apparently vegetative state was evaluated with functional MRI scanning (fMRI) and found to be able to change her brain activity when asked to imagine herself performing two distinct tasks. This study showed, at least in this one case, that a patient with no outward signs of consciousness (and therefore in what we call a persistent vegetative state or PVS) might still retain some hidden consciousness (and therefore really be in what we call a minimally conscious state or MCS).

Further, Dr. Steven Laureys and colleagues have been demonstrating that up to 40% of patients who are diagnosed as being in PVS by standard neurological exam demonstrate signs of minimal consciousness on a more rigorous exam better designed to detect subtle and intermittent signs of consciousness. They recommend this exam be used routinely to assess comatose patients, which is reasonable.

You may remember Dr. Laureys from the Rom Houben case – which was tainted by the introduction of bogus facilitated communication. As I have said – that case is an unfortunate distraction from the real research that is going on by Dr. Laureys and others. But it has successfully distracted and confused the media and by extension much of the public.

Now Martin Monti, Steven Laureys, et al have published a new study in the New England Journal of Medicine that reveals the results of an examination of 54 patients in a coma with functional MRI scanning (fMRI). They replicated the fMRI study of the comatose woman discussed above with some variations. They asked comatose patients to imagine a motor vs a spatial task, which in healthy controls causes different parts of the brain to become active, which the fMRI can detect. The ability to reliably demonstrate appropriate brain activity in response to verbal commands to imagine these two tasks was interpreted as a sign of consciousness.

Here is a summary of the results:

– Of the 54 patients, 23 were in a PVS and 31 MCS

– Of the 54 patients tested, 5 were able to demonstrate consistent responses demonstrating consciousness.

– Of the 5 patients that demonstrated consciousness, 1 was in a MCS and 4 were in PVS. However, on reexamination, 2 of the 4 patients in PVS were found to have subtle clinical signs of consciousness and were therefore in a MCS by clinical criteria.

– Therefore, of the 23 patients originally diagnosed as being in a PVS, 2 were reclassified as MCS based solely on the results of the fMRI study.

– Of the 5 patients showing fMRI results, 1 was further tested and found to be able to use the two states (imagining a motor vs spatial task) to answer 5 yes/no questions with 100% accuracy (a sixth question gave ambiguous results).

– All of the 5 patients who showed signs of consciousness suffered from traumatic brain injury. None of the patients who suffered from anoxic brain injury (from lack of oxygen) or brain infection showed signs of consciousness.

These results replicate and extend the previous work, and further support the notion that fMRI can be used to more accurately classify comatose patients as either PVS or MCS. The study design seems solid, and I like the fact that they asked at least one patient yes/no questions with known answers. This suggests that these patients are not just responding to the verbal stimulation without higher level conscious awareness. However, it is still unclear how these patients are experiencing their diminished consciousness. Perhaps they are in a dream-like delirious state.

What are the practical implications of this study? The media is focusing on what I consider to be a very minor part of this study – the one patient who was able to answer yes/no questions. This is interesting in terms of how to interpret the study, but hardly practical for communication. The fMRI machine costs millions of dollars and is not portable, and the technique of examination is very elaborate. This is not practical for anything other than one-time or very rare communication with a patient – not routine communication.

Dr. Allan Ropper (not involved with the study) observed that this technique might be most useful in communicating with patients about their care – specifically whether or not they wish to be kept alive. This could indeed be very useful – once we are confident enough that the communication with this technique is autonomous and competent.

The technique is also useful for prognosis, which greatly helps in decision-making about care. The 49 patients who had no signs of consciousness on this test may truly have a hopeless prognosis (although this needs to be confirmed with follow up, and studies specifically designed to assess long term prognosis). Being able to tell families that even advanced techniques cannot reveal any signs of consciousness can be reassuring in a way.

It is also interesting that all the patients who showed signs of consciousness were in a coma due to trauma. This makes sense, as trauma is likely to damage some parts of the brain more than others, and therefore there may be some preserved brain able to generate consciousness, even if the person is completely paralyzed. While anoxia resulting in coma is often a diffuse process (which is what happened to Terri Schiavo) – every brain cell suffers from lack of oxygen. We may therefore be able to use this technique to develop better prognostic factors based on type of injury or other parameters.

Finally, as I have also discussed before, this technique may identify patients who are eligible for experimental treatments, like implanting electrodes that stimulate the brain and increase consciousness.

This research is very interesting and encouraging, but it must be emphasized that this is still a research tool only – not ready for routine clinical use. Also, it is completely unclear what the nature of the consciousness in these individuals is. It may ultimately be a distinction without a difference. It should also be emphasized that we are still talking about a small minority of comatose patients (5 of 54 in this study).

But I hope that soon such technology will make my job easier – so that I can give families more accurate and reliable information about their loved-ones who are in a coma. Perhaps it may lead to some comatose patients being able to direct their own care. And of course ultimately it may lead to a treatment for some patients that may give them back some measure of meaningful consciousness.

10 responses so far

10 thoughts on “More on fMRIs and the Comatose”

  1. I saw this on a christian news blog with the title “Study published that assures that PVS people have some consciousness” (in spanish) with a photo of Terri Schiavo.

  2. sowellfan says:

    I read through the NEJM article, but I don’t see any mention of blinding the doctors to the answers that they were ‘supposed’ to go. It seems clear that this application of fMRI could be useful, but how reliable are these results, for indicating that the docs were getting through to a conscious person inside of that damaged body?

  3. @sowellfan
    It says, in the methods section, that “the nature of the questions ensured that the investigators would not know the correct answers before jugging the fMRI data.”

  4. KathyO says:

    @ Daneel

    Well, that didn’t take long. I wonder if families would accept that their loved ones really were gone and not coming back if this technique were performed and showed no signs of consciousness.

  5. serv44 says:

    i really appreciate this article. I saw the most basic version of this finding on CNN.com and a friend pointed me here for additional info.

    I can think of no better definition of Hell than to be in a MCS and it not be recognized.

  6. HHC says:

    Its useful to have more rather than less information on the comatose. Personally faced a crisis this past week involving a traumatic brain injury to an elderly in-law. Unfortunately, elderly patients do not recover from an accelerating fall as young patients can. Though the hospital staff attempted to use a drug induced coma to facilitate recovery, there were no real signs of consciousness involving cortical functioning. It seemed to me that independent of the full life support equipment, the patient was in PVS, and the neurosurgeon agreed. There would not be any functional living beyond the traumatic accident. I appreciated the powerful words of the neurosurgeon in this ICU room prior to his final diagnosis. He emphasized that there were no gods in the room, and asked if anyone present felt they were one. No one present felt that way.

  7. inconscious says:

    This study was a tremendous improvement over the previous study published in 2006 in terms of decreasing doubts of the validity of the “imagery” methods used. Also, it’s interesting to see some media outlets, such as CNN, actually mention the skeptical backlash to the facilitated communication business/fiasco from months ago (which, of course, they helped create). As a burgeoning researcher who plans on studying patients with disorders of consciousness (DOCs) it’s also nice to see studies by Laureys’ and Owen’s lab come out of relative obscurity. Of course, the whole facilitated communication thing hasn’t helped, as the idea of there being “no such thing as bad publicity” doesn’t work well with science.

    Hopefully, though, the publicity that’s been generated will lead the US advancing policies and infrastructure that allow for research/advancement of aid to this patient population (for example, a national patient registry of patients in a given DOC like the one in place in Belgium).

    I have to admit, Dr. Novella, that having met Steven Laureys, making plans to visit his lab and learn some things, and being a likely future researcher in his field that some of your criticisms of him seemed harsh at first. Once I realized this, I found that the experience was, overall, a good exercise in removing myself from friendships/emotional attachment to really get at solid skepticism. I guess that’s really the point of why I read your blog! Bravo!

  8. BKsea says:

    In a recent Science News article, fMRI researchers were able to show that a dead salmon was minimally conscious:


    Really a warning that some fMRI techniques are a little fishy.

  9. kvsherry says:

    I would hesitate to call this a use for allowing patients to determine their course of treatment. As it stands now, a consent is not valid if the patient has received narcotics or Ativan prior. These are adequate safety measures.

    A simple yes/no question is much easier to comprehend than a complex question like “Do you want to continue on like this?” In my mind, the fact that these patients had a severe TBI immediately puts into doubt their ability to fully analyze problems at an adult level, and unfortunately, I don’t see how a battery of yes/no questions can prove just what their functional cognitive level is.

    Do you have any ideas as to how we might go about proving the autonomy and competence?

  10. HHC says:

    bksea, Enjoyed the excellent Science news article,” Trawling the brain”. Its useful tool yo understand a past Neurologicablog post which showed a Chicago psychologist was a trying to map a personality factor to a fully conscious individual brain. The exceedingly complex statistical comparisons for the fMRI is beyond the current capacity of that psychologist’s knowledge about radiology.

Leave a Reply