 All right. Good afternoon. Hi there. My name is Toss Cochran. I am the director of Neuroethics at the Center for Bioethics here at the medical school. Welcome to the Neuroethics seminar series. This particular seminar was planned before my time, but boy, is it brilliant. It's going to be some of the top people in the field of neurology, consciousness, and ethics, talking about a cutting edge technology that promises to have some important implications for patient care. And so before I introduce our speakers, I would like to thank our funders and our co-sponsors, the MindBrain Behavior Initiative at Harvard and the Harvard Brain Initiative, are important funders for us. And the International Neuroethics Society provides funding that allows us to live stream this and post the videos afterwards, which is important. So thank you to them and thank you to anyone who's joining us via the live stream. If you are following the webcast, you can tweet questions and comments to us at HMS Bioethics is our handle and I encourage you to do so. Our co-sponsors are listed here, but in the interest of time, because I know the discussion will be sort of the most important and engaging part of the evening, let me jump in and introduce our speakers. I'm going to try and introduce everyone all at once, which would be difficult to remember everybody, but I'm not going to do them justice anyway. So let me sort of put them out on the table and then let them take the floor from here. The first person who's going to speak is David Fisher, who's one of our fourth year medical students who in large measure arranged this conference and who knows a lot about the field, has published in the field, and he's going to give us an introduction to disorders of consciousness and the new technologies that we're going to be talking about tonight, especially functional MRI imaging in disorders of consciousness. Let me talk about our three speakers. So our first speaker isn't here yet. His airplane was held up on the runway in New York, but reportedly it landed at 330, so I expect he'll walk in at any moment. And that is Joe Finns, who is the William Davis Professor of Medical Ethics and Medicine at Wild Cornell Medical College. He's the co-director of the consortium for the Advanced Study of Brain Injury at Wild Cornell and Rockefeller University. And he recently published a book, came out in August called Rights Come to Mind, Brain Injury Ethics and the Struggle for Consciousness. I've read it over the weekend. It's a very important book. It's very well done. I highly recommend it. And you know, I'm not just flattering him because he's not here yet to hear me do that. Instead of Joe Finns, our first speaker will be Joe Giacino, who is the director of rehabilitation neuropsychology and disorders of consciousness program at Spalding Rehabilitation Hospital and an associate professor in the Department of Physical Medicine and Rehab here at the medical school. He was a co-chair of the Aspen Work Group, which was responsible for developing the diagnostic criteria for the minimally conscious state, which you'll hear about. And the co-lead author of the Mohawk Report, which was an initiative to establish recommendations for lifelong care of patients with disorders of consciousness. He currently chairs the vegetative state and minimally conscious state guideline development panel at the American Academy of Neurology. And so we'll be influential in terms of how we care for these patients in the future. And he's currently investigating fMRI techniques for investigating visual and linguistic awareness or consciousness in patients with disorders of consciousness. So Joe Giacino was going to speak first after David's introduction. And then finally, Jim Bernat is the Lewis and Ruth Frank Professor of Neuroscience at Dartmouth Medical School. He's the director of the program in clinical ethics at the Medical Center, and he previously served as assistant dean at the Dartmouth Medical School. He served for 28 years on the American Academy of Neurology's Ethics Law and Humanities Committee with 10 years as the chairman. And he, in every relevant sense, wrote the book in my field, the book is called Ethical Issues in Neurology. It was one of the first most important books I read in ethics. And it's a must have for anybody interested in the ethics, in the intersection of ethics and neurology. So thank you to all our speakers. And now I'm going to have David come up and tell us about consciousness and imaging. All right. So thanks, Dr. Cochran. So, yep, my name's David. I'm a fourth year medical student here at Harvard. And there are people here with much more interesting and informed views than my own. So I'll be brief. But I just want to give you kind of a general overview of the types of disorders that we're talking about and the kinds of problems that we're faced with with these disorders. So when we're talking about disorders of consciousness, a kind of logical place to start is asking, well, what is consciousness? Obviously, that's a complicated question that we're not going to be able to answer tonight, let alone in this introduction. But it's good to have some kind of general framework to think about it. So consciousnesses can be thought of in at least two parts. One is arousal or wakefulness, just being awake. And the second is awareness of the self, of the environment, essentially the content of one's experience. And we think about, for the most part, we think about awareness as relying on arousal, in that you need to be awake, first and foremost, in order to be aware of the things around you. Okay. So what are the disorders of consciousness? So there's a few kind of main types. First is coma. So in coma, you patients lack arousal and therefore also lack awareness. How can you tell that these patients lack arousal? Well, for example, they don't open their eyes spontaneously, they don't open their eyes to any kind of sensory stimulation. And so essentially, they have really no signs of consciousness in either sense. After that is the vegetative state. These patients are awake, for the most part, but lack awareness of their self and the environment. And for that reason, this is also called the unresponsive wakefulness syndrome. When this state persists for over a month, we call it the persistent vegetative state, which is also a term that you may have heard. So after the vegetative state, there's the minimally conscious state. Now, the minimally conscious state is similar to the vegetative state in the sense that these are patients who are awake and have some kind of impairment more specifically in awareness. But unlike the vegetative state, they do still have some semblance of awareness. So it might be minimal, it might be fluctuating, but they're still aware in some sense. So that means that essentially the distinction between the vegetative state and the minimally conscious state is, is there any conscious awareness at all in the patient? And you can imagine that that's an important distinction for family members, for medical management. It matters to know, are these patients actually aware of what's going on or not? So that leads to the next question. Well, how do we determine if a patient's aware? That's a subjective experience. Well, mostly you do that on the basis of behavioral demonstrations of awareness. What are behavioral demonstrations of awareness? Well, we have some criteria to help us with that. Following simple commands, making gestural or verbal yes or no responses, making intelligible verbalizations, and any kind of purposeful, non-reflexive behaviors, like smiling or crying in response to emotional stimuli, reaching for, touching or holding objects or following a target with one's eyes or fixating on salient stimuli, these are all behaviors that are thought to signify some kind of underlying level of conscious awareness. And when they are done in a reproducible and kind of sustainable way, we think that probably these patients are at least minimally conscious. Now, there's a problem here. So when they're, when these behaviors are present, we say that likely that's a minimally conscious state. When they're absent, we say, well, probably that's a vegetative state, meaning that these patients probably have no conscious awareness at all. But there's a danger in that, which is that the absence of evidence isn't necessarily the evidence of absence. And what I mean by that is that there are reasons why you might not have any of the behaviors on the previous slide while still being conscious. So for example, you might have sensory impairment, motor impairment, speech impairment, cognitive impairment, there might be medical equipment that makes it hard for you to interact, or your levels of awareness might just be fluctuating, and the doctors, you know, or the healthcare professionals just happened to catch you at the wrong time. And so in fact, there are studies that suggest that an estimated 40% of patients in a vegetative state, patients presumed to be entirely unaware, actually do show subtle behavioral signs that in fact they are aware. Now, that's a behavior, that's a kind of issue in its own right. But it also raises a deeper question, which is that, well, what if you're a patient who's conscious but you have no way of demonstrating it behaviorally, which is plausible, right? Well, that would be missed with these kind of behavioral techniques. And so we get into another possibility, which is what if we use neuroimaging to help us. So there was a study in 2006 where a patient who for the most part was unresponsive and therefore considered to be vegetative, was put in an fMRI, functional MRI machine, and asked to imagine either playing tennis or imagine visiting the rooms of her house. Now, when healthy people do this, if they are asked to imagine playing tennis, they'll activate regions of their brain that are associated with movement. If they're asked to imagine walking through a house, they'll activate parts of their brain associated with spatial navigation. And in fact, this patient could do this essentially just as well, suggesting that this patient could actually voluntarily modulate their brain activity, even though there was no way of looking at them really, that they were consciously aware. And that there are some pretty interesting implications from that. It would be a pretty terrifying experience if you were a patient who was conscious but had no way of demonstrating it. And in fact, there was a study that followed up on this kind of study that used this method to have a patient who was also essentially non-communicative, unable to answer yes or no questions, who could use these kind of imagery tasks to answer yes or no questions. So they would say, okay, I'm going to ask you these questions. If the answer is yes, imagine playing tennis. If the answer is no, imagine walking through your house. And this patient could accurately and reliably answer these questions using this technique, even though they were not able to do that behaviorally. So these findings obviously have very interesting kind of implications. And you can imagine that there's lots of promises for this type of technology, but also a lot of potential pitfalls. So without further ado, I'll bring up our panelists. And I guess we'll start with Dr. Bernat. Did we decide? Dr. Giucino? Okay. So thanks, David. I think a very suitable introduction for where we're going to go. I have to turn on. Oh, yeah. So I think my role here is a little bit of a stage setter for the real speakers for the evening, Drs. Bernat and Finns, fingers crossed I'm Dr. Finns. So I'm going to be the guy with the boots on the ground in the sense that so my team is carrying out this work, some of whom are here. And this work is in collaboration with the functional nor imaging lab at Brigham and Women's Hospital. So we've been working together on this project for most of the time I've been here. So approaching five years. So what I want to do is to tell you a little bit about this work that you just saw a little bit about in the review that the brief review that David just did. But what I want to do it, I really want to do is to kind of begin to highlight some of the issues along the way, but really then start to move us towards some of the tricky issues that come up when we're engaged in this type of work. So this is our study. It's funded by the, that's actually a name change. So it's now the National Institute on Disability, Independent Living and Rehabilitation Research. It's a division of HHS. And so this study is entitled, as you see, Looking for Consciousness, a novel functional nor imaging approach for detection of visual cognition in patients with severe TBI and disorders of consciousness. So you just heard the story from David. Why are we doing this? Because we know that some percentage, nobody quite knows what percentage of individuals who are believed to be unconscious are actually conscious. So we want better techniques than bedside examination allows to be able to detect those individuals because maybe then we can, we don't, those individuals who are conscious, we don't know how conscious they are. So we have ways to at least now begin to tell us about the prospects for consciousness when behavioral signs fail us, but we really don't have a sense of where they, how high up the ladder they go. Why is this problem so significant and so persistent? And that is the problem with misdiagnosis. So sometimes we just don't investigate carefully enough. A bedside exam takes 10 minutes. The routine tests are given a few commands, a few prods and pricks here and there and on to the next patient. So that's probably not adequate. Behavioral signs are often misleading. Dr. Fins has arrived. We're all signing it with great, welcome Joe. Behavioral signs can be misleading. So David had a slide talking about all the potential underlying unrecognized sensory motor cognitive language problems that get in the way. And so it's a very difficult task. So some of these, some of the variance if you will is contributed by the examiner and some by the patient. What are the consequences? Well, so I've been looking for data to support this top premise, which I know to be true, but it's very hard to get data to support it. And that is, if an individual has a diagnosis of vegetative state in the neuro ICU, for example, their prospects of getting to rehabilitation, even though they're very early post injury, are less than if they don't carry that same label. Again, you'd have to sort of accept my word because I can't provide evidence for that. But I believe that to be true. Also, a misdiagnosis, meaning misconsciousness, made delay use of interventions that could otherwise promote recovery and could even lead to inappropriate or harmful interventions, in some cases, withdrawal of care, which we get very concerned about in the neuro ICU setting. So you saw the two studies that sort of started the search for consciousness away from the bedside through functional neuroimaging, the Adrian Owens group in the science, and then Martin Monti, who was part of Owens group, who pushed forward and developed the communication paradigm. So I'll just say another word, couple words about the Monti paper. So once establishing that in healthy volunteers, by asking them to imagine playing tennis or imagine walking around the rooms of their house, you can, with 100% accuracy, by looking at the activation profile, figure out which of those two tasks they're doing. What Monti did was take it a step further and say, let's turn this into a communication system. So now what I want you to do is answer some basic questions. If your answer is yes, imagine playing tennis. If your answer is no, imagine walking around the rooms of your house. And show it again that healthy volunteers can do this task very well, and that naive observers can figure out their answers with very high accuracy. Monti did the study in 54 patients. About 23 of them were in a minimally conscious state, and the remainder in vegetative state found that five out of the 54 could follow the commands like in the Owens paradigm. And one of those 54 was actually correctly signaling yes and no answers. Now it turns out that that patient was in a minimally conscious state, not a vegetative state. So it isn't that that patient lacked any consciousness, but there was no way to engage a communication system with that individual except through this functional neuroimaging methodology. But here's where I want to raise one of the questions, and that is about sensitivity and specificity of these techniques. So what you don't see in that paper is that let's just take for a moment the 23 patients in MCS. All of those patients were known to be in MCS because a bedside examination done carefully had detected clear evidence, clearly discernible evidence of conscious awareness. But remember the numbers. So out of the 23 MCS patients found to have behavioral evidence of conscious awareness, only five could follow commands using the imaging paradigm, and then of course just one had communication. Now we'll put the communication aside because none of the 23 could communicate, but that's really poor sensitivity, right? So five out of 23 that were discerned on bedside exam failed the imaging test. So we miss a lot, and the danger is interpreting the negative, which we don't ever want to do under these circumstances. So I'm not going to take you through the gory detail of this study, but just to give you a sense of what's going on in this space. So we had two aims, and one was to develop a more simpler paradigm for detecting conscious awareness. So another way to get people to sort of follow simple instructions and answer yes, no questions, then using imagery. Imagery is hard. We know that it's not so easy for healthy people, let alone individuals with severe brain injury. So we wanted to come up with a simpler task, and then we were interested in comparing the hit rate, if you will, between neuroimaging studies that look at command following and communication with the bedside examination using a standardized measure, the coma recovery scale, with the family's observations. So we did structured interviews asking the family, so do you think so and so can follow simple instructions? Do you think so and so can indicate yes and no answers? And so we end up with this matrix where we plug in the results of the three modalities, if you will, or ways of interrogation and figure out how much agreement or disagreement there is. So that's what we're interested in. And our subjects are 10 individuals in vegetative states, so completely unconscious, 10 in minimally conscious state, 10 in post-traumatic and confusional state. That's the next step up. So we want to include a group that's sort of a contrast group. No question that they're conscious, no question whatsoever, but they're severely brain injured and still cognitively very compromised, so they should actually be able to perform these tasks. It would be the idea. And then of course controls. So complex paradigm, but for our purposes, all I really need you to know is when people are in the scanner, they're going to do, they're going to be exposed to three types of stimulation while in the scanner. Passive stimulation. So they are asked to do nothing, but simply to look at different types of visual stimuli. The word yes versus pictures of scenes or buildings. Why these two stimuli, because there are super selective areas in the brain that respond to words, and then a different area of the brain that selectively responds to scenes. So we want to be sure that the visual pathway is working and that the hardware for detection of a word is active, as well as the hardware for detection of landscape scenes or spatial scenes is also active. And the reason for that is because we're then going to move to the active stimulation paradigms. So now we're going to give them instructions. They wear goggles in the scanner and they're asked to either look at the word yes or to look at the picture. So that's our command following protocol. If they look at the word yes, we get activation in the visual word form area. If they look at the landscape scene, they get activation in the parahippocampal place area. And the idea is that these areas are discernible or clearly different or different enough that we can tell which thing they're doing. So if we see this activation profile, they're looking at the word yes. If we see this, they're looking at the picture scene. But then we take it to that next level. So now we say, and here's the simplified aspect of the protocol, we're going to ask you some questions. If your answer is yes, look at the word yes. If your answer is no, look at the picture. So we've turned it into an augmentative communication system that works off the same premise. So hopefully that's sort of clear. That's the idea here. We are still acquiring data on this paradigm. So this is just another look at what we are hoping to see, the differences between activation of the visual word form area versus the parahippocampal place area. So they're discernably different and should be able to tell us whether the person is following instructions correctly and then answering yes or no. And then what we do with this at the end of the day is we produce, and this is sort of material for discussion, we produce a report which details the tasks that we did, the findings that we got, and we furnished this information to whoever provided consent. So this would be surrogate consent because patients obviously can't decide for themselves whether they're consent. And so they have these results in hand and our IRB has said once you give the results to the surrogates, it's up to them what they do with them. You can't hand them off, but the family can hand them off if they want. So that's going to set the stage for some discussion later. Toss, I can stop here. So the question is next up. Okay, okay. So let me give you two case vignettes. So these are pretty close to the belt case vignettes that have come up along the way for us and have led to lots of fairly intense, very interesting debate about how to use research findings outside of the research. Now some people would simply say it's just not to be done under any circumstance. I don't know that you're going to hear that from all of our speakers today, but you might. Families will usually tell us we've run focus groups with families. And most of the time they tell us, well look, we want you to tell us what you think obviously from your research findings, but we don't want you to hold anything back from us. We want the right to use the information or not. You have to tell us about the concerns, the constraints, etc., but we want that information and we want to use it the way we think is best. So let me give you case one. So here we have a clinician who's also involved in research activities. And the clinician is treating a 20-year-old college student who's in a coma as a result of a severe TBI that occurred following a motor vehicle accident. He's three days out from the injury. So the clinical team is scheduled to meet with a family to talk about the patient's prognosis and determine how aggressive care should be. The prognostic indicators that have been acquired to date suggest a pretty high probability of an unfavorable outcome. Most likely if you apply the conventional wisdom, this person will stay in a vegetative state or at best have severe disability. The clinician is interested in getting as much information as possible to inform his judgment of prognosis and is familiar with the neurodiagnostic procedure that's still undergoing investigation as a prognostic tool, but has shown some promising preliminary results. Clinician is also aware that the new procedure remains investigational, but is considering ordering that study for clinical purposes based on the premise that many routine clinical studies lack an established base of evidence and have the same level of risk as this one, because this is a life and death decision, literally. Could this information help inform the decision? So the question is, should the clinician researcher go forward and get these, get the studies outside of the research protocol for purposes of management of this case? So I'm sure there are some people who are saying, absolutely not. Some people are in the middle and maybe some people even pushing the other side. So do you want to leave this where it is? Do you want to engage discussion? How do you want to, okay, we can stop. Provide a nice case and I think I will segue it nicely. Yep, I know that both of our speakers are going to be happy to jump on that. Okay, so we're adjusting on the fly given, Dr. Fins, you'll have to trust that I gave you a glowing introduction and that I did mention your book and that I did recommend it to the audience. Great, thank you. I'm just glad to be here because my flight was, we were taking off from LaGuardia and mid runway, they aborted the takeoff and there was an engine failure. So I'm glad to be here. Okay, great. Oh yeah, okay, we're not following up. That's why it's not reading yet, hopefully. Top, yeah, that says laptop. That says podium laptop. Okay, it doesn't actually look like it's there yet. There you go. We're close. Okay, great. We're close. Well, I'll be able to see the side of it. Okay, great. Thank you very much. Again, thanks for the invitation to be here. And what I'm going to try to do is, is, is to place a lot of what, where Joe went. Joe's comments into a sort of a bioethical context and, and, and just start off with some comment from John Dewey, who was a mid century philosopher, was very interested in pragmatism, instrumentalism, and sort of the relationship between theory and practice. And, and he has this great comment, I think, because we're here because of emerging technology that inventions of new agencies and instruments create new ends that create new consequences which stir men to form new purposes. And I think one of the fundamental issues here is that before we had this new imaging technology, we didn't have this problem. We did not know there was a discordance between what people did behaviorally and what they might be going on internally in their brain. So technology sort of opens up to this problem. And technology also interestingly might be part of the solution. Now I just want to, and I miss David's talk. I presume you're David and I presume you gave the talk. And, but, and I saw your slides, but I just want to just remind us how far we've come in a very short period of time since, since all of us old guys were in medical school. This is basically all new news. And Lewis Thomas, who was a great physician writer in 1980, wrote, you know, Nature of Conscious is a scientific problem, but still an unapproachable one. It's becoming approachable and, and worth the cusp of, of this new knowledge. But of course it creates new problems. I, I share this picture because David didn't, didn't put this, this earlier picture in his slide set for two reasons. One is that Jim Burnett and I are both on the front page of the New York Times. And it was commenting on a paper about a passive stimulation that Nico Schiff and Joy Hirsch and colleagues and Joe and Fred Plum, colleagues did it at Cornell, which showed, which for me was for me personally, a huge eye opener. This was a forward backwards language paradigm that was played to patients who were in the minimally conscious state. And these folks actually registered language in the forward direction, but not backwards. So the same frequency spectrum didn't, didn't elicit the same response. So they were processing language. That was like a real eye opener. And it happened in 2005, right at the midst of the Shibo case exploding. And, and it points out, what I said on the front page, this, this study gave me goosebumps because it shows this possibility of this profound isolation. These people are there. They've been there all along, even though we've been treating them as if they're not. And that's the moral crux of the problem. Okay. These people who were here before were diagnosed as vegetative may actually responding to the outside world. And the Venn diagram that once was the vegetative state, and we don't have time to get into all of this, there was a subset of people who were minimally conscious. Through the heroic efforts of Joe Giacino and other colleagues, we've begun to partially adapt this other category of people. And that's a difference that makes a difference. The other point I want to make in showing this slide and this slide here is that the minimally conscious state has a net, they have network responses. And the way the brain, as I understand it, works is through a series of networks. This is a study that Stephen Lorries did in, in Liège about pain stimulation and patients who were vegetative. All they do is light up the primary sensory area. They do not activate the network response. So one of the major distinctions between this category patient and vegetative patients is the possibility of network response. And we'll see how that's important. I think you've covered this with the, the Adrian Owen study, but, but I want to just put it into slightly different language about motor cognitive dissociation. Again, what patients do behaviorally is not what they're doing on the scanner. And Nico Schiff and I wrote a little piece in the Hastings Center report around that time calling this non-behavioral MCS. So back in 2007, just to give you a sense of how rapidly progressing this history is, a number of us were at a conference in Stanford, including Jim and, and, and I think that's Nico and, and I don't know if Joe, Joe was there. And, and the conclusion in 2007 was that, that these results were not ready for prime time, eight years ago. And that, and that there were all kinds of reasons to be very cautious about sharing these results with families and they were still in the research purview. And then of course this study came in as, as Joe described. And I want to make two points in addition to what he said. The first is there are real sensitivity and specificity issues. And it's the primacy of the exam, the coma recovery scale that, that Joe helped to originate was, was better than neuroimaging. So whatever we say about neuroimaging, the behavioral bedside exam in the skilled hands of a skilled neuropsychologist, and most of the people who do this are neuropsychologists, is more accurate than the neuroimaging studies. So it's kind of a utilitarian, you know, greatest good for various number. However, the neuroimaging study for that one patient had a, had a real impact for that one individual. However, as Joe pointed out, and if you read through the lines carefully, that patient turned out to be minimally conscious retrospectively after the imaging study pointed them more aggressively to the behavioral exam, but had, would not have been identified perhaps without the neuroimaging. Of course they established the communication channel. So, so, so questions, what do we do with all this new information? And, and, and as a bioethicist, we've all been schooled about this sort of strict dichotomy between research and therapy between the clinic and the research encounter. And I think this distinction sort of fades at the margins here. And drawing upon the work of Susan Wolfe, writing about incidental findings, she's written that, that, that research findings can be shared with, with individuals or with surrogates when there's an emerging view that research is bear some clinical responsibility towards research subjects. And the italics are added there. And the criteria for this is, does the information have potential health or reproductive significance? The validity of the test, and Joe was talking about that earlier, to some discrimination between the risks and benefits of disclosure and the utility of the information. More recently, Henry Richardson, who's a philosopher at Georgetown, wrote this marvelous book entitled Moral Entanglements, in which he's thinking about research ethics in the developing world. You're doing a study on, on, on, on AIDS and, and in sub-Saharan Africa, and people have schistosomiasis. And you have the resources to treat those people. Should you treat the schistosomiasis? Federally, it's prohibited by federal law to use research dollars to treat clinical problems overseas. But he makes a very compelling argument that, that, that subjects should receive some degree of clinical care based on your degree of interaction with them, their degree of need. And he really begins to erode this dichotomy between research and practice. And, and I, when I reviewed this book for the, for, for a journal, I, I related it to the needs of this population. And this population is so woefully undertreated and underattended to in the clinical context that many times the investigators who are, who are working with these people have information that no one else can get and have a, I think a profound moral obligation to provide some help and share their expertise. This is a piece that I wrote in the American Archives of Physical Medicine and Rehabilitation Medicine that begins to describe some of these problems in the book. Also catalogs this in detail. These people, you know, suffer from, you know, premature discharge, premature powder of care recommendations. They're often seen as potential organ donors before they've declared themselves. They're discharged to places without, without a credible diagnosis. Sometimes they're given a false diagnosis to ease discharge. All these things are, are cataloged in the book based on interviews with over 50 families. So I think that there's, there's a real need. And I think if we think about it, in the context of ancillary care obligations, that dichotomy between research and clinical work begins to get, get some, get some little, little grayer in the middle. So let's talk about some of these, these issues about proportionality and risk. Well, first of all, all the technologies we're talking about, FMRI or DTI, which is a structural, a structural utility, not a functional study, and PET, which looks at metabolic activity, are minimal risk or slightly more than minimal risk. And they're established as safe and routine clinical practice. We're not talking about new interventions that we don't quite understand. There are methodological risks that I thought Joe was going to describe. But since I got here late, he has described. And that really hinges on a couple of things, positive versus negative results. And because of what he described, a negative result is never just positive. And the issue is really a type two error failing to identify consciousness when it is present in, in that kind of issue. The second point is timing. This is a really important point. And, and there's a recent paper from Canada about the utility of neuroimaging early on. And one of the, one of the things that you just have to remember that these are not diagnoses. These are brain states. And it's like the difference between a fixed dementia and a delirium. People transit from one of these states to another. So when that, when people are moving from coma to the vegetative state, to the persistent to the vegetative state, and then into the minimally conscious state, it's a moving target. For traumatic brain injury, it's up to a year. Anoxic brain injury is three months. And if you get hypothermia for a cardiac arrest after anoxic brain injury, you really don't know what the outlier is because that's changed the levy criteria. So the point of using neuroimaging as much as we invest in technology, if you label somebody in a certain way early on, it could become stigmatizing. And it could be a label they carry forever. So it's very, very concerning. It doesn't mean we shouldn't do neuroimaging on these patients, but we should understand there's a contingency because they're going to evolve over time. Now, Fred Plum, this issue of contingency, Fred Plum, who many of you may have heard of, who, who identified the vegetative state with Brian Jeanette and also the locked in state and was, was my teacher, had this, this quote about, about probabilities. I think a lot of this is probabilistic. He said, you know, there are absolute versus probabilities and ethical decisions. Probably the most difficult, yet crucial point to be discussed is the one of relativism. When one can rely on the probabilities and the occurrence and reaching an ethical decision, rather than await the definitive event, no matter how unlikely. This is the point. Many ethicists hold the beliefs that are at least stated in absolute terms. By contrast, an increasing number of medical problems tend to be decided in relative terms. And Jim is also a student of Fred Plum. So the fact that two of us are here, it may be an endorsement or not of Dr. Plum's pedagogy, but that was a joke, guys. Okay. Very serious here in Boston. So the point is, we can't think about this as much in absolutist terms because it's really more nuanced than that. So how do we deal with the sensitivity and specificity issues that Joe described? I think we have to understand every single test that we use against a Bayesian pretest probability. Yes, history and physical matter. And does it make biological sense that we're kind of moving in a certain direction that somebody's got a certain kind of diagnosis? Also, behavioral metrics, when properly done, are better than neuroimaging. And that requires expertise. Also, because the minimally conscious state itself is a state where the behaviors are episodically and intermittently presented, you can't have one data point. You have to do this repeatedly over time and in different times of the day because people's diagnosis or state depends on their highest degree of function. Not what they often do, but what they do one or two times. The metaphor is it's kind of like a flickering circuit in a light bulb. The fact that the circuit can flicker means that the network is intact. Again, the intact network to activation versus a vegetative patient who's truly vegetative will never have that flicker. That light will never go on because that network is not intact. The next point is the need for nested judgment about imaging in the context of a broader sort of clinical assessment. While there Penfield, the great neurologist and neurosurgeon who founded the Montreal Neurologic Institute, his autobiography was called No Man Alone. He was really an interdisciplinary kind of guy. And I think when we think about neuroimaging, it should be no test alone. No single test will tell you the answer to your questions. How do we deal with mitigating risk and talking to families? We've got to frame it as a research result. We have to talk about contingency. There's also the issue in a lot of the surrogates that I talked to from my book and Joe's got the same data in a different cohort is that the surrogates express the right to know and not the risk aversion that often generates what we share with people. And then how do we do it? We like to use the intermediary of a physician who's treating the patient so that there is a kind of a portal for that discussion. Utility, the question of utility and the obligations of disclosure following Wolf and Riley as well. And Susan says, if there's no benefit, there's no obligation. If there's a possible benefit, it may be discussed and discussed. If there's a strong net benefit, it should be discussed on the subject or the surrogate opts not to know. In my view, and this is a quote from the book, there's nothing more important in my view than knowing that a patient may be conscious especially when there was a paucity of motor output and the possibility that neuroimaging data obtained for research or not might suggest that an imaginal thought to be vegetative might actually be aware. I can't think of anything more important in medicine than not missing that key point because it has to do with personhood, respectful persons, and whether we discard this individual or not. What happens when you tell a family that their loved one who they thought was vegetative is in fact conscious, albeit minimally conscious? Family interactions change. Parents and spouses and kids who never talk to their loved one start talking to them. There's the issue of pain management. If you know somebody's minimally conscious and not vegetative and they've got that network activation instead of what Stephen Lowry showed in his vegetative patients with the primary sensory area and they can perceive pain, then you've got a moral obligation to attend to their palliative cure needs. There's a story that always gets me about Terry Wallace who was a man who woke up in 2003, woke up after being in a coma or vegetative state for 19 years after a car accident. These are terms that were misused in the media, a lot of media attention. In fact he was minimally conscious for most of those 19 years and in early 1990s, his mom and by the way every name that I share with you, I've got IRB permission, hip approval, all those things Bob, don't have a don't have a stroke, and he were totally covered there. Mrs. Wallace told me in his story that one day in the early 90s the nurses in Terry's nursing home called up Mrs. Wallace said you got to come and see Terry. He's not right. The nurses evidently hadn't read the textbooks about vegetative patients being right or wrong or just basically inert, but they were these mothers and their mothers and motherly intuitions and they said Terry's not right. And Mrs. Wallace got to the nursing home and she saw Terry and he was kind of like he looked startled and kind of bug eyed a little bit and what had happened, the man he was in this nursing home with all these elderly people and the other bed in the room who had dementia had asphyxiated himself in his bed and died. Now we don't know what Terry Wallace experienced but he wasn't right. Ten years later, thanks to Joe's work, there's a whole chapter on Joe Giaceno and his life in the book, because of the evolution of the coma recovery scale and the minimally conscious state, we know that Terry Wallace had actually been in the minimally conscious state and had some degree of awareness at that time. So the notion of that isolation, being aware at some level we don't know, people always say, well what did they feel, what did they say when they began to come out of that state and could express themselves, they're amnestic for what happened in that period of time, but there is some awareness. So the palliative care mandate and then the last point, does it matter? Is there actionability here? The answer is yes. There is this evolution of neuroprosthetics that can help make a difference and there are three kinds. There are drugs, amantanine and zolpinum, Joe and John White did a large randomized clinical trial, was a new neuro medicine a few years ago on amantanine. Devices, deep brain stimulation, again we collaborated on deep brain stimulation as well as the imaging methodologies as a communication tool. So if you know somebody is minimally conscious, there are things that you can begin to think about doing. So there's actionability. Here's what a mom said. This is Maggie and Nancy Worthen and she's the main character in the book. And first she says, as a surrogate, it's right to know and take the risk. Let's move beyond paternalistic judgments that protect the institution and let's empower families. And on the information, on contingent information, she felt we must be informed and this is a quote, but I think it still has implications for treatment even though it's contingent. It should. It would be important if you had information. At least don't share it with me but share it with Dr. Katz who was Maggie's doc so he knows everything so that he could decide treatment and he and I could use that speculative information how we want to treat Margaret because we need it in order to decide on treatment and that's how she perceived it. The other point is getting back to this, what you can do is that the minimally conscious state is a state of potentiality unlike the permanent vegetative state. And Jose Manuel Rodriguez Delgado who was the famous neurophysiologist who did the stymicever with the bull in Spain was a Yale physiologist talks about potentiality in the brain. It's like a highway able to accommodate traffic and facilitate the exchange of visitors among the many cities. The highway, however, cannot create cars, trucks, merchandise, businessmen, workers and the life which circulates along the road makes functions possible but by itself is a useless stretch of pavement. The minimally conscious brain is that highway system. It's that network system and basically what we're trying to do is drive the cars on it and get some activity on it. But differently we want to move somebody onto this inflection point and one of the tragedies and one of the hopeful things about the minimally conscious state is the longer you're in the minimally conscious state does not predict that you're not going to come out of it. There's no time coefficient there and unlike most things in medicine the longer you're sick the worse it's going to be. If those networks are there there's that potentiality of them being activated unlike the vegetative state. Once you're in the vegetative state and it's permanent there's no return and all those people who had late recoveries from the vegetative state were in fact probably misdiagnosed. So the idea here is this is the potentiality that we're trying to go after and why this matters and what is at stake I think is fundamentally a human rights issue. Fundamental human rights issue and this is why the book is called rights come to mind that if you look at the demography of this population conscious individuals have been ignored, they've been sequestered and they're potentially salvageable. That's not to say that I'm romanticizing this brain state. Nobody would want to be in this brain state but let's meet people where they are and if they're conscious they should be recognized as such. I've tried to expand the arguments from the Americans with Disabilities Act and one of the fundamental premises of the ADA and the Olmsted decision which was the Supreme Court upholding and expanding the ADA was to maximally integrate people in civil society and people's integration depends on voice. With voice comes community and I think one of the issues when we think about consciousness and research results and research ethics is that it transcends this sort of narrow regulatory frame that we often think about these problems. I'll just skip ahead and say it's going to be complicated because if we use these techniques people may again over extend their utility to questions that people are not prepared to answer so invariably when the Monty paper came out I got asked to comment on you know should people be able to use this technology to decide whether they want to live or die and I said you know we've opened up a communication channel with this technique but in some ways it's like a very bad cell phone connection and the bandwidth is very narrow and people are certainly not going to be at the level of consent and refusal because they can't ask questions they can't interrogate us their tension can be can be wavering there's latency of response in a study that we did that was the John Barton study a lot like the Monty paper we thought people weren't responding they were responding into the next question and outside the region of interest so there's a lot of questions there so non-response is non-dispositive and it's important that as we begin to bring patients in through this technology and try to integrate their voice we do it in a responsible way and I've talked about we'll talk about this during the discussion about a mosaicism so the patient's voice is integrated in their response way let me let me give one last quote and I'll stop because I'm getting the look from from pass here and that is to quote one of your former professors who might have taught in this very room I don't know Oliver Wendell Holmes the father and I think it's he wrote a piece called the border lines of knowledge in some provinces of medical knowledge and science is the topography of ignorance from a few elevated points we triangulate the vast spaces and he says the best part of knowledge is that which teaches us where knowledge lives off ignorance beginning nothing more so clearly separates a boulder from a superior mind from the confusion the first it's a little bit it truly knows on one hand and what happens what it knows on the other and this is most true knowledge which deals with living beings I think this is a great cautionary note as we think of that this cusp of knowledge that we're on and I want to thank you for your indulgence and for your attention thank you very much thank you very much show that wasn't quite what we planned but I think it worked perfectly it where I think we're well placed now to let Jim Bernat speak about this issue in the context of his expertise and experience in the clinical realm let me see if I can figure out how to get us back where we need to be Jim you want to come on up and okay Dr. Bernat thank you great thanks very much it's a pleasure to be here can you live with them seeing the next slide yeah it's good for me because I don't know what's coming up next main PC to projector secondary PC there we go thank you thank you okay great okay so my talk today is going to be about the medical practice impact of functional neuroimaging in disorders of consciousness and the overview see if this works it's not working let's try this that doesn't work either you're actually going through it boy seems to work there oh my gosh but you can't see it now okay I think that may be what we want beautiful good thank you gosh okay we're good to go here what I'm going to talk about today is how do these functional imaging research modalities pertain to the clinical practice of managing these patients and in that consideration I first want to talk a little bit about the traditional diagnosis and prognosis as determined by the multi-society task force and admittedly these data are now about 25 years old and then focus briefly on diagnosis prognosis communication medical decision-making in the future and I have a chapter in Martin Money's book coming out about this so the multi-society task force was a group of 10 people representing five specialty societies who met and published a two-part paper in the New England Journal in 1994 that tried to summarize what was known about the vegetative state at that point and came up with the following diagnostic criteria the first group was unawareness of self and environment no sustained reproducible or purposeful voluntary behavioral response to visual auditory tactile or noxious stimuli and no language comprehension or expression and looking at this first set of criteria it's obvious that they're delineated in negatives that is what these patients cannot do and that creates some of the problem in the diagnostic realm the second part of it was that they had persisting brainstem and hypothalamic function so there were sleep wake cycles preserved autonomic function that allowed the patients if they were young and had no other comorbidities to survive with long intervals with aggressive medical and nursing care and they had intact cranial nerve reflexes so that the diagnosis then prior to the functional neuroimaging studies was to fulfill what one might call the negative diagnostic criteria about what they couldn't do and they should be examined with optimal scales and it's been mentioned here Joe's coma recovery scale revised which is the state of the art for determining evidence of awareness on a neurological examination and other things that we do at the bedside such as having the patient gaze at himself or herself in a mirror in a hand mirror and see if there's any flicker of recognition interviewing nurses and caregivers sometimes family members will say that there is some evidence of responsiveness and I know I've been called to the bedside and sometimes I can see it when they demonstrate and other times I still don't see it and I'm not sure that it exists repeating examinations at different times in different days these people have fluctuating levels of neurological function and some days there's really nothing and other days there may be more responsiveness it's been pointed out that the false positive rates and considering the minimally conscious state so this was the state of the art clinically prior to the first paper by Adrian Owen and his colleagues in science I wanted to just mention one other recent paper because it wasn't mentioned by the preceding speakers and this was a paper published last year in Lancet that looked at fluorideoxyglucose PET scanning and comparing its positive and negative predictive value to functional MRI most of the studies have looked at fMRI and they looked at fDG PET as well in a cohort of 126 patients a very large series for disorders of consciousness 41 of whom were vegetative 81 of whom were minimally conscious and forelocked in syndrome and then they correlated the findings on these neuroimaging studies to the clinical findings on the carbon recovery scale revise and found that the PET sensitivity for diagnosis of MCS was 93 percent with 85 percent congruence with the CRSR whereas fMRI in the same group the sensitivity was 45 percent and congruence 63 and that in the vegetative people one or the other showed evidence of responsiveness in about a third of them so for diagnosis it seemed as if the fDG PET was a little more sensitive than the fMRI so the question is now with the introduction of this fMRI with diagnosis that there are a few patients who fulfill the clinical criteria for the vegetative state but who show awareness behavior on either fMRI or fDG PET and therefore they're not minimally conscious by definition because I mean they are minimally conscious they're not vegetative by definition because vegetative people don't have any evidence of awareness as has been pointed out by previous speakers false positive diagnoses of VS are common and result from inadequate examination and I think in my experience some of them are many of these are traumatic brain injury patients who had been on neurosurgery services they were clearly vegetative during the hospitalization but during hospitalization or transfer to a rehab center they've gradually recovered to minimally conscious but that wasn't really noticed because they weren't reexamined in a very careful way and because of the negative delineation of the vegetative state criteria which as was pointed out invites this kind of false positive error and the question then should diagnostic criteria be expanded to include fMRI and fDG PET data probably in the future and we'll come back to that now let's turn to prognosis and looking back now at ancient history on the multi-society task force what they said we I was part of that was the prognostic data that in non-traumatic most of these were hypoxic ischemic neuronal damage from cardiac arrest some stroke patients some encephalitis patients that in the non-traumatic people if there was no evidence of awareness of cell for environment at three months the probability of that occurring subsequently was very small whereas in the trauma people in order to achieve the same degree of diagnostic accuracy we said you had a wait a year probably longer than that but that's what we said and we recognize that there were a few late recoveries and that the popular media stories of full delayed recoveries were misleading I'm not going to have time to get into the mortality and that business so what's happened to the old prognostic information in the current era of functional neuroimaging this was pointed out that that atrian Owen in his group their vegetative patient that was said to be vegetative the original science 2006 report a patient and then others than Martin Monti at all described in the New England Journal in 2010 the ones who seem to show the so-called willful modulation also was that subgroup that seemed to be destined for clinical improvement as Joe Finns just pointed out many people who are vegetative particularly after traumatic brain injury are in a transitional phase where they're going to spontaneously improve to at least a minimally conscious state and it seems like the people who show the willful modulation by functional neuroimaging are the ones most likely to be in that state who will then develop clinical evidence of awareness as Owen's patient did it 11 months now it's important I think when we look at clinical syndromes like vegetative or minimally conscious state these are syndromes they're not diagnoses they have a spectrum of severity they don't imply a pathophysiology and it varies a lot the two big groups will be the traumatic brain injury and the hypoxic ischemic group from cardiac arrest and the stroke groups well it turns out that those have very different natural histories as the multi-society task force pointed out the hypoxic ischemic neuronal damage people have the worst of those prognoses and that that the ones that have the evidence of willful modulation are largely restricted to the traumatic brain injury mechanism and they are not largely the other non-traumatic forms like stroke or hypoxic ischemic neuronal damage now there have been a few of those but almost all the others are traumatic brain injury and it's important I think to get out of the syndromic diagnosis and into specific mechanisms because they have different outcomes so back to the standard study they looked at prognosis and compared the fdg pet to fmri and their group of 126 patients that I mentioned and it turned out that the fdg pet outcome prediction accuracy was 74 percent whereas the fmri was 56 and about a third of the so-called vegetative patients were found to have awareness by one the other or both and of that subgroup 69 percent gradually recovered clinical evidence of awareness so that I think shows again that the fdg pet seems to be prognostically have higher positive predictive value for improved clinical improvement subsequently and if you chose the willful modulation that's the subgroup destined to improve spontaneously uh both prior speakers have mentioned the communication issue and the one patient of mandi that was taught to use yes or no and remember this picture now everyone has shown or adrian owens picture with the tennis imagery and the spatial navigation and the idea that one patient who was taught successfully to think about tennis for yes and spatial navigation and walking in the houses of room for no was able to have a communication system that was accurate but this is a very unusual circumstance despite the fact that there are now quite a few reported people who can do the willful modulation and the question is why and the answer seems to be that there are serious physical barriers to language production understanding and production in the subgroup of patients and that is they have a high incidence of aphasia because of language dominant hemisphere damage that occurred in the traumatic brain injury many of them have global cognitive impairment from diffuse brain damage of varying types uh there's sedation for mannequin bolson drugs and other medications that are given and their medical comorbidities involving heart lung liver disease kidney disease that produce metabolic encephalopathy that further depress brain function and the totality of all these factors leads to the outcome which is that the ability to use this language to establish a reliable communication system is a very rare phenomenon indeed but we should try to do that um a word about medical decision making we like to have medical decision making patient centered and not position centered which is the old way of doing it and that the willful modulation with communication potentially allows patient to communicate by participating although that's rare as I said but one question it raises is whether a yes no response what I would call a binary response is sufficient to create a valid consent for treatment or refusal and the best model for addressing this question is our locked in syndrome which is much more common and we have a lot more experience in trying to communicate with these people in the old days before we had computerized systems that had laser beams on the corneas and allowed people to do more sophisticated forms of communication we were left with only vertical eye movements or eye blinking and of course people like Jean Dominique Bobie in his book the diving villain butterfly was able to dictate that by eye movements but for the most part yes no answers don't give the richness and the nuance that we need to make clinical decisions particularly regarding whether someone should continue to be treated or not so I would argue that although this is exciting with using willful modulation by fmri that we need to be a little cautious about what we do with that information so in summary the impact of fmri and fdg imaging is that patterns of cortical activation may suggest the presence of awareness behavior in some patients in whom the neurological examination does not even done in an optimal way and the same patterns can be predictive of prognosis that is identifying that subset who is destined spontaneously to have a higher probability of improving and that therefore they may become useful clinically and make that transition from their current status as research tools into a clinical usefulness but before we do that we need more detailed studies with greater numbers of patients to establish true predictive positive and negative predictive values to get more uniformity and the paradigms for doing this we need more reliable funding this is a group of patients that are not there's no mechanism for getting a lot of funding for them a lot of the funding for the research is from private endowments and we need better and more reliable funding this will allow us to map the anatomy and physiology of human consciousness to revise later and i think we will and maybe the next five or eight years revise the diagnostic criteria of vegetative and minimally conscious state to incorporate these types of studies and i'd be interested to hear from joji asino and your work with the development of the new diagnostic criteria for this where you think this belongs and of course improving technology of communication for those few patients who retain that capacity so thank you very much so i'm going to have our speakers come up to the table and grab microphones if they could and david's going to man the microphone and take questions from the audience but i i think dr. jasino set us up with the first question that i think will touch on most of the aspects that are in play here imagine that imagine dr. bennett dr. finns that we've got a patient who is in a is in a coma shortly after a traumatic brain injury we've got this research protocol available we could use this protocol outside of the study without much extra expense or extra time and we could potentially use it to inform the clinical care for this patient and dr. jasino set us up with a yes no maybe and i'd love to i'd love to hear everybody's perspective on where they come down on that so could you just restate the final part of the question and yes no maybe um if you've got a you've got a patient in whom you could use your available research protocol to to to study them using fmri and look for evidence of awareness that's not clinically evident now and it won't cost anybody anything significant to do it and the question is are you permitted to do it are you obliged to do it are you forbidden from doing it and on what grounds yeah i think it's worth first mentioning that the number of institutions that have this capacity is very small it's very different than things like ordinary mri where you put the patient in one end of the machine and the images come out the other and they're all relatively uniform and we know what to do with them we know how to read them these are experimental paradigms that are designed on site they've gone through a lot of sophisticated processing and clever um technology to to get them and they're not exactly exportable so first of all the probability that someone would be admitted to a place where they could even do this is pretty small so but let's take your question and take the minority probability that okay well someone is admitted say to um new york hospital new york presbyterian hospital now um and uh where they do have this capacity uh then i think that a strong case could be made that uh with proper consent about what it might mean that that this could be done i think that anytime you're taking something and this is the whole nature of translational research that is basically a laboratory technique and you're considering using it there should be some learning that uh occurs from it and whether it be an n of one trial or some other way of using that information i think it should be it should be used but i think it would be hard if you had the capacity and the situation existed not to um to do that but i think that it should be incorporated into some kind of a learning research methodology i agree i think but i i think we have to be very cautious by being seduced by the technology and and not and not and i would rather have fly joji asino up for a fraction of the cost and have you do a good exam because we as we've seen that's that's that's more predictive and it's better there are also so many confounders in the acute setting that that you could you could really get a result that could be very misleading and and these people just carry these diagnoses with them and it and it makes it may make it very difficult for them to get insurance they get labeled as vegetative or not and in fact it's not it's not a permanent in a state that they're in so i think um so i would be wary because i'm not sure it's going to add very much there's a there's a potential of a harmful label um and the information uh can be obtained in a better way with more careful clinical work uh but we tend to jump to the technology quickly so i would say um you know if it's part of a protocol if it's well understood but i wouldn't do a one-off because i don't think it's really helpful it's not to say we shouldn't do structural studies because that can be very important um but i'd be very wary um uh because i think it it doesn't really add very much and it could be horrible could i ask a follow-up question given that you're one of the centers that studies these people um would that be a would there be a possibility for them to be we don't we wouldn't do it you know we probably wouldn't do it in that context we would be on a protocol we don't have a protocol for that the people that come to us earn a more chronic state so and let me just make this a little trickier so well i think i can say that while it's true that these protocols are not running on every hospital's scanners in some places and let's assume now that we begin to see diagnostic centers pop up so places specialize to do this so if you go to mgh now the imaging package that standard incorporates a lot of these measures so the or these sequences so they're not special in that setting anymore and that becomes very interesting now so you'll have fmri you'll have dti and lots of other sort of advanced sequences as a standard part of the protocol now for imaging so think about that but here's the thing so the flip side of what families say and remember the scenario that was painted was the clinician is also the researcher in that example and is you know left with this difficult situation of discussing prognosis with the family so families actually also said to us sort of out of the other side of their mouths that they were not comfortable with clinicians necessarily being furnished with the results of investigational studies because that information could one of two things could happen it could enter the thinking of that clinician so if the study was a negative they can't undo that and so that's clinician is less potentially less invested or changes the path of care to some extent because of that even if consciously they would say they would never do that that they couldn't prevent that from happening and then the other which was a very practical issue was the possibility that a clinician would write a note in the chart that would then be audited by an insurance company who then saw the negative findings reported and that might have impact on future authorization of care i'm going to open it up to questions for a second so dr. fins I appreciate your defense of a good clinical exam but you seem to be walking away from the worth in case that you describe in detail and the very good rationale that the family give for wanting the findings of neuroimaging to be entered into right the understanding of the state of their family member the big difference was that that this was that particular patient was diagnosed as vegetative when she came to us her neurologist here in Boston wasn't sure that was true and the neuroimaging was was contextualized against a really good set of clinical evaluations and it was kind of like a tiebreaker the other point which is really important it was nowhere near the acute stage it was it was in the chronic stage and so I think it's a difference that really is important about where you are in the timeline because you know she was already in either the permanent vegetative state or in that kind of minimally conscious state where she had yet to emerge with reliable communication so it was a different context all right we actually also have a question from twitter so we're going to answer this question from twitter thank you so this is a question coming in from our online audience from philip kelmeyer of the university of fryberg so a lot of the focus has been on fmri assessment of consciousness but what is the role of current or future for electrophysiology and especially when we talk about communication as you did does fmri based communication have a future giving its low bit rate and practical constraints or should we start to think about brain computer interfaces and if we start to think about that you know what do you know these electrophysiological based devices pose in terms of new ethical dilemmas in these patients so hi philip i know him so this was a this was a controversial uh discussion a few years ago damien cruz adry known wrote a piece that was in the lancet uh purporting to use uh eeg as a way to identify conscious in our group led up by nico ship john victor uh andy gold fine we disputed the findings and we thought it did they didn't make biological sense and it was kind of a signal to noise problem and uh there was an exchange of letters and you guys can read uh read them and decide whose side you come out on but one of the issues that i think is important is that that the barrier to entry for eeg is so much lower than this technology that if you're wrong you've cloned a huge mistake so we've got to get it right do i think that eeg eventually will play a major and does sort of play a major role and there's other work that is very very constructive i do think it will play a major role but again it's the same issues about tests uh specificity and again it has to be contextualized it's going to be no single test by itself is going to be able to give us these answers it's really is a it's going to be a kind of a portfolio evaluations that's going to help us understand whether somebody is in one of these brain states when the exam the clinical exam doesn't tell us already i think we're going to get our answer much more quickly because these studies are are being completed at a much faster rate than they are for neuroimaging the paradigms are very similar more flexible you know using evoked potentials so but i think they're fraught with all the same difficulties that fmri is i just think that we will have more data faster to figure out the what the sensitivity specificity is the one maybe exception to this is uh the group in malon who's coupled um you guys probably know about this the coupled uh tms with eeg and developed an index called the perturbational complexity index and basically what they did was a pulse of tms to perturb the brain and then used eeg to track the spatial and temporal trajectory of the signal and skipping all the details were able with a hundred percent accuracy to sort individuals using this pc i this which is a mathematical index to sort individuals who were conscious from those who were not this was a very impressive study um interestingly the the the it couldn't work out gradations of um consciousness so in other words it treated mcs patients who were very low functioning only had visual pursuit the same way as individuals who were basically just about communicating uh reliably but when you and and they were able to do this with um controls under anesthesia um people asleep and awake and then individuals who were unconscious i.e vegetative or comatose versus those who were minimally conscious emerge from the minimally conscious state or locked in um that has not been replicated yet but this is a very well designed study that um could change the game a bit all right so we have a couple more minutes so one more brief question so i guess to go even even a little bit simpler than uh you know fm rye eeg or tms eeg has anybody pitted this against just very strict clinical exam in other words we're going to have three different docs examine you every single hour multiple times per day you know for every day in a row for a month and what percentage of vegetative state patients then show signs of conscious awareness when you do that and how does that compare to you know these people that were now examining with fm rye or eeg or tms eeg so yes and no yes in the sense that the gold standard if you look in the literature is consensus based or team consensus based team diagnosis so the studies that you saw cited the schnaker studies which are probably the best and most recent um compared the teams consensus diagnosis against these other measures so the but but you also asked that you pose the question very specifically you know they're looking at this patient over a series of multiple studies in the same day over seven days that answer is no nobody has done that um there's at least one paper out now that is that does look at the fluctuations in the course of a day um which and it becomes very interesting because if you were to use one of those exams for diagnostic purposes you're going to get the exam changing literally in some cases hour by hour so this point about the transience of the state is really important i mean that said there's really strong evidence that the diagnosis and joe made a really good point before which i think shouldn't be missed and that is the diagnosis is the peak exam you can't accidentally get up there you can go down you can you know sort of down regulate for many different reasons and often but you can't accidentally get to a point where you're clearly following commands so that that really is the i think what the the what what should be used as the indicator but the question is how many exams do you have to do to capture that and that we don't know the answer to yet it's worth mentioning since we're talking about functional neuroimaging at this discussion that the same fluctuations occur in that that occur in the clinical examination and that if you just did one exam at one point it's going to give you one piece of information but if you repeated it six hours later or two days later it may be another just as there is fluctuation in the physical examination so it just adds to the complexity of how these one of the reasons the bold might be picking up on this is you have to wait sorry you have to wait 10 seconds after you ask the question just because of the delayed nature of the hemodynamic response and so you have to sample out of these patients the bedside how often do i wait 10 seconds for them to move their eyes one direction or another after ask the question you usually ask the question if i don't see an eye movement within a few seconds i ask my next question or so which is my mirror which actually might may account for why the CRS performs as well it has a fixed 10 second response window now how did we get to 10 seconds pretty damn arbitrary but we standardized that and for 25 years it stayed the same it's you know not longer than that it's not shorter than that i don't know if it's right we actually wrote a grant to see if it was right it didn't didn't get funded but but i think that that's another really good point here so you know when we do these bedside exams so what you're getting at now is how you interpret a response so what's the window within which one should wait before you decide yes that's clearly there or that's not the CRS standardized and that and many other things so i think that that may be it's crowning glory all right so we're out of time thank you so much to our panelists for coming thank you all for coming as well um yeah round of applause yeah