The President's Council on Bioethics click here to skip navigation


FRIDAY, June 25, 2004

Session 6: Neuroscience, Brain, and Behavior V: Deep Brain Stimulation

G. Rees Cosgrove, M.D., Associate Professor of Surgery (Neurosurgery), Harvard Medical School, and Attending Neurosurgeon, Massachusetts General Hospital

CHAIRMAN KASS:  Some wag in the room at the break indicated that we need to develop a new kind of disorder for the DSM which is called change the question and quick to metaphysics disorder.


And we have fMRIs ready for all of you between now and the next meeting.

I don't want to take any time away from the session.  The change in the weather has led some of our colleagues to have to leave before this session is over, and on their behalf, I offer their apologies for the necessity of leaving before we're done.

It's a great pleasure to welcome Dr. Rees Cosgrove to the Council.  He's Associate Professor of surgery and neurosurgery at Harvard Medical School and the Associate Visiting Surgeon at the Mass General Hospital.  He kindly interrupted his vacation to come back and offer us a presentation on the just newly emerging uses of deep brain stimulation not for motor disorder, but for disorders of behavior.

And, Dr. Cosgrove, thank you very much.  Welcome, and we look forward to the presentation.

DR. COSGROVE:  Dr. Kass, thank you very much for inviting me.

What I would like to do briefly this morning is give a very short historical perspective because I think that's paramount to understanding some of the moral and ethical issues that are involved with surgery for psychiatric illness; briefly describe for you the current practice of ablative surgery for psychiatric illness; then discuss the issues of deep brain stimulation and some of the very vestigial or rudimentary, early experience, and it is tiny, of deep brain stimulation for psychiatric illness, specifically obsessive compulsive disorder; and then trying to address some of the ethical issues which Dr. Kass so kindly directed me to consider; and then leave plenty of time for questions and discussion.

The modern era of psychosurgery was begun by this man, Egas Moniz, who is a very celebrated and famous Portuguese neurologist who experimented by injecting alcohol into the frontal lobes of 20 institutionalized psychiatric patients and thought that 16 of the 20 were favorably improved.

He subsequently went on to devise a more discrete operation in the frontal lobe through burr holes, and this was such a major public health problem in those days with the asylums full of the psychiatrically impaired and mentally ill that these initial, early experiments in treating psychiatric illness were very favorably received because, in fact, they did actually improve behavior, and Dr. Moniz was, in fact, awarded the Nobel Prize in medicine in 1947 for his work in this area.

And he was the man who coined the term "psychosurgery."

At the same time, the champion of this field in this country was a psychiatrist-neurologist named Walter Freeman, and he, in conjunction with Washington, D.C. neurosurgeon James Watts, performed multiple prefrontal lobotomies, which was disconnecting the entire frontal lobes with the use of a sort of calibrated butter knife inserted through holes in the coronal temporal region and inserted to the midline.

Dr. Freeman himself was an unusual man.  There may have been some psychiatric diagnoses potentially attached to him.


DR. COSGROVE:  But his zeal and his sort of overenthusiastic adoption of this procedure really was difficult for the neurosurgeon.  It's rare that the neurosurgeon is the responsible character in these teams.

But actually the neurosurgeon showed great responsibility by actually declining to participate and collaborate with Dr. Freeman because he thought Dr. Freeman was over extending the applications and misusing the surgery.

That didn't really stop Dr. Freeman who was a neurologist remember, who then devised a procedure that he could do himself, and this was the famous transorbital "icepick" procedure in which a sharp blade was inserted over the globe, over the orbit through the very, very thin roof of the orbit into the underside of the frontal lobes, and he performed thousands of these.

He actually would cross the country in his van, really advertise his arrival in major metropolitan centers, and actually perform these at asylums and hospitals throughout this country, and he'd perform ten or 20 in a morning and then off he'd go.

So Dr. Freeman was probably in large part responsible for some of the negative feelings toward psychosurgery of this sort of closed, nonstereotactic methods that were used.

As you might imagine, these procedures were associated with some significant mortality and morbidity.  It's estimated that there was about a ten percent major mortality and morbidity.  Nevertheless, these procedures were considered actually useful, and despite the fact that the National Commission on the use of Human Subjects and behavior in research experiments in the mid-1970s said that at least half of the patients whom were operated upon sustained benefit from these procedures.

The psychosurgery had its grand demise slowly throughout the '70s for many reasons.  I think in large part the most compelling ones were these moral, social, and philosophical aspects in which I think we heard a little bit this morning about operating on the brain to heal the mind.

Issues arose probably in large part because of, again, the morbidity associated with these gross and very crude techniques.  There was a lot of, I think, people harmed by the early times of surgery.  There were a variety of political stances against this kind of surgery because of the outrage, and there are a variety of medical and legal issues.

But probably most compelling was that in the mid-1950s, the first psychopharmacological agent, chlorpromazine, was introduced.  And so right at this time alternative psychopharmacological agents became available.

And as we've heard, it's far easier to give a pill than do an operation to treat illness.  then over the next, you know, 30 years a huge variety of more selective psychopharmological agents that were very effective in treating schizophrenia, depression, and even obsessive compulsive disorder arose.  And so for a large variety of reasons, surgical interventions for psychiatric illness basically declined to just a handful of cases throughout the world.

And if you want to read a great book to describe all of the somatic therapies that were used in these times of great need, read Elliot Valenstein's book.  It's a wonderful document of this history.

Currently, however, surgery is still practiced in rare occasions, and the only two indications that we typically perform the surgery for are major depression and obsessive compulsive disorder, and the surgery is only performed in those patients who have severe and incapacitating psychiatric illness.

The degree of severity is typically estimated or estimated with a Beck's Depression Inventory score of greater than 30, and the global assessment and function score of less than 50.  These are people who are severely ill and completely incapacitated.

In terms of obsessive compulsive disorder, we typically only operate on patients who have a YBOCS — that's called the Yale-Brown Obsessive Compulsive Score — of 25 to 30, and this is an enduring illness.  This is chronic illness usually of many, many years' duration.

In addition, we only perform the surgery on patients who are being completely refractory to all forms of conventional therapy.  So you must look at this as a salvage operation or a palliative procedure, and it's only performed on patients who have actually exhausted all forms of modern psychopharmacology and pharmacotherapies.  Typically this means that in the obsessive compulsive disorder group that they've had three trials of modern SSRIs with up to maximum tolerated doses augmented with either lithium or Wellbutrin or clonazepam, any of those things.

In addition, one of the major therapies for obsessive compulsive disorder is behavioral therapy, and they have to have exhausted all forms of behavioral therapy or at least committed to 20 or 30 hours of behavioral therapy, and they've also had to fail when appropriate electroconvulsive therapy, which we know is a very practical and important intervention for major depression.

This procedure is undertaken, and I will review for you our own.  We have all patients who are referred to us undergo evaluation by a psychiatric neurosurgery committee, and this is a committee that has been in existence for the past 20 years at our institution, and it is composed of six members, three psychiatrists, the former chief of psychiatry in our institution and then a specialist in obsessive compulsive disorder and a specialist in major depression.

There's one neurologist, myself as the neurosurgeon, and a recording secretary to document all of the information that passes through our hands.  and it is this expert multi-disciplinary panel that is charged with the selection and implementation of the interventions.  And this panel, primarily the psychiatrist actually, are responsible for insuring that the accuracy of the psychiatric  diagnosis, the adequacy of drug and pharmacological therapies, the adequacy of behavioral therapy and ECT. 

One of the psychiatrists is assigned as the primary on the referral and actually does a review of a detailed psychiatric referral form, and all of the records are reviewed and summarized for the committee, and then all of these things are discussed in a committee, and there has to be unanimous approval by all members of the committee that the patient meets criteria for surgical intervention, and there's a whole bunch of other tests, including EEGs, MRIs, PET scans, neuropsychological testing.

And then if they meet these criteria, then they're brought in person for evaluation by the primary psychiatrist on the case, the neurologist and the neurosurgeon for final decision making.

One of the important aspects of modern psychosurgery is the use of appropriate outcome measurements.  What we have attempted to do and what is occurring now in the past decade has been implementation of these outcome measurement scales.  These are the same scales and using the same thresholds in terms of determining successful treatment as are used in pharmacological drug therapies.

So one of the important things is if we're going to promote any sort of surgical intervention for a psychiatric illness, we have to use terms and outcome scales that are recognized by the psychiatric community, and so that we can show by comparison how they rank with appropriate psychopharmacological therapies.

These are very standard and accepted throughout the world.  A Beck's depression inventory or a Hamilton depression inventory of 50 percent improvement from baseline would be considered a success in the psychiatrist's eyes.

Obsessive compulsive disorder, which we'll talk about a little bit more, is much more difficult to treat, and there are fewer successes.  And so in this instance a 35 percent improvement in their YBOC score is considered a successful pharmacological intervention or behavioral therapy intervention.  A global assessment of function is sort of a psychosocial level of functioning and the minimum is a 15 point improvement in the GAF.

And then also, although a subjective score, this clinical global improvement scale is a seven point score with one or two being either very much improved or much improved.

So these are the scales by which modern psychosurgery is measured, and in terms of our own institution, we actually put additional — because we're looking at individual patients and not groups of patients, we actually characterized our outcomes as a patient who responds to our intervention as having in the depression scale either a 50 percent improvement in their Becks and a CGI of very much improved or much improved, or in the obsessive compulsive patients, a 35 percent improvement in their YBOCS and very much improved or much improved.

And of course, patients have a continuum of response, and so we considered partial responders as meeting the numerical criteria for a pharmacological therapy or either that or being considered very much improved or much improved by the rater.

And then all other patients were considered nonresponders, even though there might be some improvement overall, but they didn't reach significance or threshold.

So if we use these very much more stringent criteria, these are very much more stringent criteria than was ever used in the older psychosurgical literature, and I think that's in large part explained by some of the differences in outcomes.

And one also has to consider that, in fact, in the old psychosurgical literature none of the SSRIs and current modern pharmacological therapies were available so that the patient on whom we're performing this surgery on are much sicker, in general, and have failed a whole host of selective pharmacological agents.

However, these are two of our own studies.  This was the first prospective study ever done to look at cingulotomy, which is one of the procedures, ablative procedures, that is performed for obsessive compulsive disorder, and performed prospectively using unbiased, unrelated observers, and those more stringent clinical outcome criteria, and in our group of those patients who had failed everything else, you see about a third of the patients became responders and, you know, 17 percent were partial responders for an overall response rate of about 45 percent.

And we subsequently continued this prospective accrual of data, and so some of these patients are in here obviously, but more recently with a larger number of patients, with a longer follow-up, surprisingly almost identical response rates.

So now neurosurgeons typically would look at a response rate of 30 to 45 percent as being not particularly encouraging.  If we had a 30 percent response rate or success rate in surgeries that we do, we wouldn't be doing much surgery anymore.

But I think that psychiatrists, if you take that this is now a complete salvage rate, these are patients who failed all of other forms of therapy, and I think that if the psychiatrists in the group said that they did a drug trial in which a new agent was added on to everything else that was being done and they got a 45 percent response rate, that would be a powerful new drug in the treatment of obsessive compulsive disorder, to salvage a completely treatment refractory group.

So while these numbers are not fabulous, they are, I think, impressive nonetheless.

Now, that's one particular procedure.  That's cingulotomy in modern times.  The other typical procedure performed for obsessive compulsive disorder is capsulotomy, and in this instance, this is the gamma knife capsulotomy results from the Brown Group.  Unfortunately it's unpublished results, but this is a very impressive group with a lot of experience in dealing with severe and intractable obsessive compulsive disorder, and they have a similar number of patients with a similar degree of follow-up, and the gamma knife capsulotomy is done with radiosurgical lesions in a slightly different part of the brain, in the anterior capsule of the brain bilaterally.

And what's interesting is using, again, appropriate criteria for rating outcome, they have 22 out of the 35 patients responding, so for a 63 percent response rate.

What is very similar in terms of the two kinds of ablative surgery performed for this condition is that there's no immediate benefit from intervention.  In fact, it goes six to 12 months before we begin to see improvement, and in fact, as we follow the patients further and further, in fact, the success rates go up, and that's true for gamma knife capsulotomy.  That's true for cingulotomy.

So as we follow the patients out further, they improve more, which is completely in contradiction to the natural history of obsessive compulsive disorder and argues against any sort of placebo response.

So if these ablative interventions are so successful, why would we want to consider deep brain stimulation?  Well, there are a variety of reasons.  Deep brain stimulation is now currently widely applied to the treatment of movement disorders, and so many groups in the country are very familiar and expert in the technology.

But the real advantage of deep brain stimulation is that it's reversible.  What is done is that using stereotactic techniques and the same techniques that are used to make these small lesions in the brain, instead of making a lesion, we implant an electrode with multiple contacts, usually four contacts, into the target zone.

And so because it's reversible and we're not creating a lesion, any side effects associated with implantation or stimulation can be dialed down or you can turn the stimulator off, and the side effects and the benefits are reversible.

So this allows us to explore areas that would not be previously conceived as possible to place lesions in.  The best analogy is subthalamic nucleus stimulation for Parkinson's disease.  No neurosurgeon with experience would want to place a lesion in there because the target is so small and the real estate so expensive surrounding this small, you know, five millimeter nucleus that any minor error in lesioning could create a devastating and irreversible neurological deficit.

Now, by placing an electrode into the area, we all do now with great regularity and with great safety.

So that is a primary advantage of deep brain stimulation.  The other thing, it's adjustable.  So one can adjust in terms of getting therapeutic benefit, and one can adjust in terms of any negative side effects, and it also is adjustable potentially over the course of that patient's illness.

So whereas a lesion is succinct and defined and irreversible, deep brain stimulation is adjustable, which has very specific advantages.

We've talked about how it allows placement in otherwise risky targets in the brain.  So, in fact, most targets in the brain now are potentially accessible by deep brain stimulation.  It's familiar technology to us, all neurosurgeons and stereotactic and functional neurosurgeons.

And the other important thing — and this is where I think it creates certain ethical issues for the Council — is that it reduces psychological barriers to implementation.  There's something about creating a small lesion in the brain that neurologists, psychiatrists and lay people and patients have a problem with, although it has been used successfully over the past 50 years.

But if you talk about stimulating the brain, and the fact that it's reversible, the barrier to considering this kind of intervention drops significantly.  I can tell you that that's true both in the Vegas nerve stimulation study for depression.  That's a relatively low risk procedure, and patients would volunteer.  They'd come into the office with their neck exposed like this and say, "Can I be part of this trial?" because it's a relatively low risk, and it's stimulating.

Similarly, when you stimulate the brain, I think a lot of negative biases naturally are reduced.  I'm not necessarily saying that's a good thing, but it does actually reduce these barriers to referral and barriers to implementation.

Deep brain stimulation, as we know, is currently accepted and has FDA approval for all sorts of treatment of movement disorders, intractable tremor, Parkinson's disease, and the dystonias, and certainly in pain. It's widely performed throughout the world for these indications.

It's under investigation for intractable epilepsy, cluster headaches, and obsessive compulsive disorder, and soon depression.  So these are still experimental.  We don't know the results of these studies yet.

But there is the potential for a wide variety of behavioral and other psychiatric conditions:  anorexia, morbid obesity, addiction, self-mutilation, violence and aggressivity, and schizophrenia.  All of these, in all of those indications, ablative surgery has been performed in the past.

But you realize that here's an interesting moral question.  Anorexia nervosa is a life threatening condition.  We've all seen cases of that.  Now, the psychiatric or the psychological situation is that the patient is uncomfortable with their body image typically.  They can look in the mirror and they can say even though they are thin and what you and I would say that's a very thin person; they think of themselves as ugly and fat or their perception of themselves is fat.

If you did an operation that restored their willingness to eat and consume, because many of these patients in their worst conditions are actually so malnourished that even minor physical activity can actually result in pathological fractures.  They die of starvation.

And so here's an interesting question.  If you take that patient, to save the patient's life and you do an operation to make them consume more calories and eat so that now we're happy with the image of the patient, but now they're looking in a mirror and saying, "God, I thought I was fat then.  Now, I'm" — it would be pure psychological torture for the patient.

So that's an interesting conundrum.  Even though you're doing the best thing that you think for the patient and you might save the patient's life, that's a horrible position to put the patient in.  Anyway, something to think about.

Now, there have been approximately greater than 25,000 cases of deep brain stimulation performed worldwide for Parkinson's disease pain, tremor, dystonia, a variety of things.  As of May of this year, there have only been 23 cases performed for psychiatric indications.  So you're talking about .001 of the experience.  So this is in its absolutely embryonic stages.

And yet so why are we interested in it?  We're interested because of the early results and because of some of the advantages of deep brain stimulation which I've spoken about.  The first published experience was from Belgium by Nuttin and they operated and used the anterior capsular target.  So instead of making a lesion placing electrodes into this area, into the exact same area in which we would perform a capsulotomy.

So we know that capsulotomy empirically has a pretty favorable track record in refractory OCD patients, but instead of making a lesion, we say let's stimulate, and in four of his patients, three of them became responders and meeting criteria of that same sort of 35 percent improvement in their YBOCS.  One patient was a nonresponder.  And this was done using blinded observers and was done in a very appropriate fashion.

So that was the initial experience that prompted enthusiasm in this area, and I should say that one of the reasons that it's important to discuss deep brain simulation is that there's so much enthusiasm in the area that, in fact, some of it may need to be reigned in, and the reason I say that is very year I give a seminar on psychosurgery for neurosurgeons, and you know, typically I tell people it's not something that you want to do very much of because, first of all, it's not paid for.  It's all gratis.  It's uninsured by Medicare.  So all of the professionals and the hospital institution gets no money for Medicare patients.  It's one of the eight specific procedures that Medicare and CMS denies.

But whenever I ask a group of neurosurgeons, stereotactic and functional neurosurgeons who do surgery for Parkinson's disease, do surgery for epilepsy, do deep brain stimulation, I ask, you know, "Who's considering doing this this year?"

The hands go up, about 75 percent of them, and that to me is a scary thought because it's a neurosurgical procedure, but it's based upon an expert multi-disciplinary group of psychiatrists and neurologists, and the neurosurgeon actually in this instance is the technician, skilled technician nevertheless.

But in any case, so this was the original paper.  There have been a couple of anecdotal reports, and this initial experience has prompted a 15 person/patient investigational trial in this country, the results of which are not yet available.

One of the things that was learned from this early anterior capsule experience was that although it's nice for us to think that deep brain stimulation, you can turn it on and off and, therefore, you can have a control state to this.  Patients and evaluators, both the patients and the evaluators know when the DBS is on.  So you really cannot blind this.

In fact, the Belgian investigators sent in medical students to try and figure out if they could accurately predict whether they could tell whether the stimulator was on or off, and they were 97 percent accurate in determining whether the thing was on or off.

So thinking that you can do one of the .- one of the proposed advantages of deep brain stimulation is that you can do blinded studies.   You can't do blinded studies.  Patients and evaluators know when the stimulators are on or off.

PROF. SANDEL:  How does the patient know?

DR. COSGROVE:  Perception, feeling, how they're feeling better.

PROF. SANDEL:  No, but independent of the relief of the symptoms, do they know it's on?

DR. COSGROVE:  No, because these thing are turned off and on with a magnet, and it's simply their perception of how they're feeling when it's on.

DR. HURLBUT:  How does the observer know?

DR. COSGROVE:  Well, I think it's actually how in talking about the patient and looking at their behavior, looking at their movements, looking at their interaction.  It's a behavioral thing.

The other way, of course, you have a monitor which you check.  This stimulator, the electrode is attached  via a subcutaneous lead to a pulse generator here that's placed just under the clavicle, just like a heart pacemaker is, and you can interrogate this through the skin, and you can program and change the settings all through the skin, just like .-

PROF. SANDEL:  With a magnet?  How do you do it?

DR. COSGROVE:  It's a little computer that's telemetry, via telemetry, the same way you do cardiac pump pacemakers, the same technology.

And the neurologist or the psychiatrist, as we do for patients with Parkinson's disease, you can select which contacts, the current, the pulse duration, the frequency, and you do that all through the little hand held device that is superimposed over the pulse generator.

PROF. SANDEL:  Who holds the clicker?  Like where is that, in the doctor's office?

DR. COSGROVE:  The doctor has that, yeah, yeah.

The patient can turn the device on or off with a magnet that they just pass over the device.  So they only have the ability to turn it on or off.  The physician is the one who has the ability to program.

So what's very clear, and this gets to your question, is that when the DBS is on, patients were more alert, more spontaneous, less anxious, and less depressed, very clear.

When the stimulator is on, it does not appear to impair frontal lobe tasks.  So neuropsychological tests looking at frontal lobe tasks, decision making, all of these things.

The improvement in their Yale-Brown obsessive compulsive score is sustained if DBS is continued, but when the battery fails or you turn it off, the YBOCS improvement goes back to where they were preoperatively.  And then you turn it back on again and you can see the sustained improvement again.

PROF. SANDEL:  Is it immediately or over a period of time?

DR. COSGROVE:  No, it's over a period of time.  What is immediate is the mood effects, but the obsessive compulsive traits or disorder symptoms do not improve right away.  They take time.

And the other interesting thing is that it requires high stimulation parameters, and what happens is what or when.  So it means you go through a lot of batteries, maybe every one and a half to two to three years you have to replace the battery.  So not inexpensive.

And the other very interesting thing is that the optimal effect is contact dependent.  So we have four contacts in there, and Paul Cosyns, who is one of the investigators in Belgium, relates this very wonderful anecdote that one of the patients who successfully treated has, you know, their four contacts, and she says, "Well, Dr. Cosyns, when I'm at home doing my regular things, I'd prefer to have contact two, but if I'm going out for a party where I have to be on and, you know, I'm going to do a lot of socializing, I'd prefer contact four because it makes me revved up and more articulate and more creative."

So there's an example of contact dependency, and we have our own patient who is a graphic designer, a very intelligent woman on whom we performed the surgery for severe Tourettes disorder and blindness resulting from head tics that cause retinal detachments, and we did this in order to try and save her vision.  The interesting observation was that clearly with actually one contact we could make her more creative.  Her employer saw just an improvement in color and layout in her graphic design at one specific contact, when we were stimulating a specific contact.

And that raises this very thorny issue of improving performance with deep brain stimulation, which is not where we're at obviously, and I don't think it's a place where we should ever go to, but you can see that different stimulation through different contacts has different effects.

So one of the things that Dr. Kass asked me to try and address is are the indications for deep brain stimulation for neurologic disorders, such as Parkinson's disease or epilepsy; are they different from psychiatric disorders?

And it's my belief, and it's actually the belief of all members of our committee that, in fact, trying to differentiate surgical intervention for Parkinson's disease from psychiatric disease is artificial.  Trying to distinguish those things is artificial.

I believe that psychiatric illness is a manifestation of the mind, the disease of the brain and disease of the mind.  So if we choose to perform surgery to help a patient, it should be done more based upon the assessment of the risks of your intervention versus the possible benefits that can be supplied, and then a critical element obviously is the ability of the patient to give informed consent.

And I think that one of the important things for this committee to understand is that the patients on whom we're operating, patients with severe depression, the patients with severe obsessive compulsive disorder, these are severely ill psychiatrically, and yet they are in the vast majority, they are completely aware of their illness and they are completely able to give informed consent.

And so one has to avoid being paternalistic about our protection of patients in the sense that if they are aware of the risks and benefits and if they are able to give informed consent, then we should allow them the opportunity to explore some of these novel interventions in the same way that we would allow them to explore enrollment in a clinical drug trial.

So to me the differentiation between neurologic versus psychiatric indications is artificial and unnecessary, and I think it stems in large part from  the history of psychosurgery in the past and some of the hold-over from the early and less precise interventions, but it's clear that outcomes have to still be measured using the clinically validated rating scales that I have told you about.

And is deep brain stimulation different from pharmacological trials?  Many times we're dealing with the same patient population.  Most of our patients who come to surgery have already failed multiple novel investigational pharmacological agents, and there clearly is a difference, however, from a pharmacological add-on agent to a surgical intervention, and it has to do with the surgical risk.

The risk of deep brain stimulation for intracranial hemorrhage is probably somewhere between one to two percent.  The risk of neurologic deficit associated with that is not insignificant in that one or two percent.  The risk of malfunction of the device, infection, abscess, all of those things, disconnection, a whole bunch of different things, is probably in the five to ten percent range now.  So it's not trivial. 

There are also risks of adding on pharmacological agents, of course.  You could get an idiosyncratic reaction and have a problem, but you know, it's much easier to stop a medication than to explant a surgical device.

So that's a major difference between deep brain stimulation and pharmacological trials, but the design of these trials and the selection of the patients, the follow-up, and the usage of appropriate outcome rating scales should be similar.  The only difference, I guess, is some of the selection criteria.  I would propose that, in fact, for surgical interventions patients should be refractory to all appropriate therapies, whereas one might consider drug trials in patients with mild depression and you use a different agent or, you know, moderate depression or moderate OCD.  I think surgery should be reserved for the most refractory and severe cases.

And then the issue arises.  You can certainly do placebo blinded crossover trials with drugs.  There's some intimation that you could do this with deep brain stimulation, but I think that's fallacious.

I think that, first of all, you can't really do a placebo surgical procedure, not a good one.  You may have discussed some of these, but certainly there is some micro lesioning effect of just placing an electrode onto the target.

As I've said before, you can't really blind the patients nor the observers.   You can certainly do crossover trials with deep brain stimulation because you can turn it on and off, and I think that that's a real value.

I'll finish up shortly.

One of the important issues and outstanding issues for deep brain stimulation is that we are not clear what the optimal targets are.  We don't even know what the optimal stimulation parameters are, and we don't know what the long-term effects are.

We don't understand how deep brain stimulation works.  We know that in gray matter it typically inhibits gray matter structures, and in white matter it typically excites white matter structures.

But any neuroscientist knows that even in gray matter nuclei, the deep brain nuclei, that yes there's a predominance of nuclei or neurons, but they're all attached with a myriad of white matter tracts, and so it's not as simple as we make it out to be.

It's extremely expensive in terms of time and money.  The hardware, the equipment is expensive, and when you're utilizing one of these pulse generators every one to two years, that's a very important annuity to the manufacturer.

Not only that; it's extremely expensive in terms of time for the treating physicians.  Seeing the patients, adjusting the stimulators, we know this for a fact in the Parkinson's group.  Always trying to get at a little better.  There's no "okay.  that's good enough," and accepting it as such.

And you can imagine in the psychiatric population of always trying to get that little bit better so that it's endless.

One important issue is to realize that currently, you know, deep brain stimulation is device based, and there's only one manufacturer in the world.  So Medtronic has a monopoly on this.  Medtronic is the sole financial supplier for the investigational studies.  Most of the people who are involved in these studies are Medtronic consultants in some way or another, or if not directly, indirectly receiving a lot of research support for their activities from Medtronic.

Now, I know many of the people, and they're all upstanding people who are involved.  I know the people involved in the OCDDBS study, and they're all upstanding and wonderful people, but one has to be cognizant of this very fact, that there's no alternative to the supplier.

And one of the other big issues, and it's true for all psychiatric research, is that there are no good animal models exist.  So we are doing this.  We are doing experimentation on humans.  In Parkinson's disease we had great animal models.  We understood some of the basic neurocircuitry, and we had hypotheses in which we directed our interventions, and we don't have that in deep brain stimulation for psychiatric illness.

So I'll conclude by saying that deep brain stimulation for psychiatric illness and indications are currently experimental and remain so.  The preliminary experience appears encouraging and, therefore, proceeding thoughtfully and cautiously seems appropriate, but — and this is the big "but" — we must guard against the indiscriminate and wholesale experimentation and repeating the mistakes of the past because this is, I think, an important opportunity to do this right.  We won't get a third chance to do this right.

On a very final personal note, I will say that as a neurosurgeon — and Ben will back me up on this — as neurosurgeons, we see an incredible amount of death and disability in our daily life, in our daily work, from a huge variety of different illnesses.  So we're used to seeing death and disability and suffering.

And in my near 25 years of doing neurosurgery, never in my professional experience have I ever witnessed the suffering that these people have, and it's suffering, constant suffering, and such disability that affects not only the patient, but every member of their family and friends around them.  It is horrible.

So that anything that we can do to alleviate or lessen, reduce these patients' suffering is important work.

Thank you.

CHAIRMAN KASS:  Thank you very much.

There are a couple of our colleagues who are going to have to promptly leave soon, and I'd like to give them the first shot if they would like to take it, to ask a question.  We're going to excuse them shortly.

Michael Sandel is one.

PROF. SANDEL:  This is just a naive factual question.  Is this different from or the same as what we read about in the popular press about the use of high powered magnetic treatments for depression?

DR. COSGROVE:  No, this is different.  What you're talking about is transcranial magnetic stimulation, and this is quite different from that.  There is a possibility — it does supposedly the same things internally in the brain, but noninvasively, but it's not continuous.  You go and receive the transcranial stimulation.  It stimulates the train in certain areas and thereby that's how we think that the improvement might result from.  It's electrical stimulation.

But it's only done episodically, and once these electrodes are placed, it's internal and continuous.

PROF. SANDEL:  And do you have a view about the comparison between the two, the promise of the two?  How would you compare them?

DR. COSGROVE:  Well, the great promise of transcranial magnetic stimulation is that it's completely noninvasive, and I think that the problem is that I think that without continuous stimulation, I think that it's less likely to be successful on the long term, but both areas are investigational, completely investigational.

PROF. SANDEL:  Right, and there's no data really to compare them yet.  It's still being developed?

DR. COSGROVE:  No, there's no data.  One synergy that might come from this is actually predicting which patients might respond to deep brain stimulation because if you can focus your stimulation in a certain area, then you might be able to predict which patients would respond, and then you would go to the risk and effort to implant an electrode if you could select that.

This person seems to be responsive to transcranial magnetic stimulation.  Therefore, why don't we take the next step and do deep brain stimulation?

Thank you.

CHAIRMAN KASS:  Before Michael goes, it seems to me, just to try to connect this conversation with the one we have in the previous session, admittedly this is at a very early stage and there are obstacles to having this implemented because of the Medicare restriction and the like, but Dr. Cosgrove listed amongst the potential applications of these addiction, self-mutilation, violence and aggressivity.

And assuming that one had rather severe instances of those things that this is not a treatment of first resort, but for refractory cases, would you be uncomfortable following this as a mode of treatment?  Would this be making us guilty of some kind of wrongful understanding of the underlying foundations of a disease if in the retail business you've got a patient with a severe problem of this sort, and Dr. Cosgrove through deep brain stimulation can rescue them?

PROF. SANDEL:  Well, yes to the first question and no to the second.  Yes, I would be uncomfortable, but only because I'm squeamish about all of this stuff, but no to the second question.  I wouldn't rule it out on moral grounds without knowing more about what kind of success it could achieve.  I wouldn't say it's some fundamental violation of our humanity or of our understanding of moral responsibility, no.  I'm squeamish about it, but that may —

CHAIRMAN KASS:  And squeamish about it in a way different from being — are you equally squeamish about doing this for Parkinson's disease?

PROF. SANDEL:  Yes, it's just as squeamish about planting the thing into the brain, but morally, no.  Morally I don't see a difference, provided it —

CHAIRMAN KASS:  So the fact that it's a behavioral disorder as opposed to a motor disorder is no part of your concern?

PROF. SANDEL:  No, for reasons that were partially developed in the earlier discussion.  I don't see the intervention to deal with behavioral disorders as crowding out or being inconsistent with other interpretive or therapeutic ways of treating or of understanding.  I don't see the two kinds of description as incompatible and at odds with freedom in a way that on a certain picture they would be.

CHAIRMAN KASS:  Questions?

DR. McHUGH:  But I would be.  I would agree completely.  I would say yes and no in the same ways, but I would be squeamish for another reason than just invading the brain, and that's the reason that Dr. Cosgrove brought up.  Whereas in Parkinsonism we do understand the mechanisms that we're working on in relationship to these things; we have no clue as to what the mechanism is.  Now, I'd still do it, but I'm squeamish until we begin to discover whether, for example, doing these brain stimulations are releasing certain endorphins or exciting the pathways for self-stimulation and things of that sort and then I would become less squeamish because I would then know what we're doing.

Because we don't know what we're doing here, even though we know that it's effective, I would be squeamish, but I would be accepting of it.

PROF. SANDEL:  Well, I would agree with that.  That doesn't reach the philosophical question that you were asking though.  Yeah, I would accept that.

CHAIRMAN KASS:  Dan, do you want to comment?  You look like you were.

DR. FOSTER:  Well, just one quick question.  You've mentioned over and over again how expensive this is and so forth.  Let's say if it were fundable in some sense.  Would you have a ballpark figure of what, let's say, the first two years of treatment might be?

DR. COSGROVE:  That's a difficult figure to come up with.  The equipment ballpark figure is in the $15,000 range.  That's for the first time.  Each time you put in another IPG it's $8,000.

Hospital cost for the initial implantation is probably in the $30,000 range, and again, these are Massachusetts rates.  So they're very low.

But then the real hidden cost is all of the time and multiple visits required for stimulation adjustment and that.  It has got to be over the space of — I can't really give an accurate number on that because I haven't been involved in the trial, but it's significant.

DR. ROWLEY:  But doesn't that have to be weighed against the costs?  Otherwise, these are not patients who otherwise you're spending no money on at all.  So it's the differential that you're asking about that's important.

DR. COSGROVE:  Yes, that's absolutely true, but I guess one has to also remind people that nobody is ever cured with these operations.  I think that's very important.  So that nobody ever gets off medication and nobody is ever cured, especially in the OCD population.  Their function may be improved and restored more back towards normal, but in the patients on whom I have operated on and what we have seen, nobody is ever cured and they remain under treatment, both pharmacological and behavioral and medical.

DR. ROWLEY:  But can they go back to work?

DR. COSGROVE:  It's very rare because they're so severely ill and they've had such disabling illness for so long.  We don't know about that in the deep brain stimulation because we're talking about a tiny experience.

In my experience with ablative surgery, yes, there are some people who go back to work.  There are some people who are working still.  Those are the minority, but, yes, there are a percentage that will go back to work.

PROF. SANDEL:  Why don't you play out a little bit the objection that underlay your question before?

CHAIRMAN KASS:  It wasn't so much an objection.  It's partly what interests us here is whether any kind of initial disquiet that anybody might have about these kinds of — in fact, Mike asked before:  is there a difference?  People would be happy giving drugs for these things but might be unhappy doing brain surgery.  Is that just because there's surgical risk or is there some kind of quasi philosophical reason that you don't somehow fix people by putting your hands on the brain?

And if that's your concern, how do you differentiate between going to work on Parkinson's disease or removing a tumor and actually doing this, even if we don't exactly know what we're doing?

PROF. SANDEL:  So we say no, but you may think yes.  So could you say a little bit, or what do you think?  What would you say?

CHAIRMAN KASS:  I don't have a firm opinion.  In other words, this is a real question.  And, indeed, the more one sort of talks about this, the harder it is for me to make the distinctions between.  That's partly why I was leaning on you in the other session.  I'm not sure it's so easy to make a distinction between a behavioral disorder.  The lines are fuzzy.

DR. KRAUTHAMMER:  I'm not quite sure it matters whether we know the mechanism or not.  When we first started using anti-psychotropic drugs, I'm not sure anybody knew the mechanism, or lithium.  I don't know the entire history, but a lot of these there's an extremely obscure, but it worked, and if it worked it works.

I'm not sure that's the salient.

I wanted to ask our presenter.  Do you feel personally any differently, a difference between ablative and stimulative surgery?

DR. COSGROVE:  No, because the scientific evidence isn't there for deep brain stimulation, you know.  So there are theoretical advantages for deep brain stimulation.  It's an important opportunity to learn more about how the brain works in these disease states and how it may be modulated.  So I think it's an important moment, but I'm not necessarily sure at this point that deep brain stimulation is better than ablative surgery, and we won't know that until we actually get many more years of experience.

DR. KRAUTHAMMER:  If I could just ask you, you had expressed some concern about the enhanced creativity by your patient who used electrode number four.  Could you draw you out on that a bit?  Assuming you have a stimulative treatment that helps someone and they do have enhancement of normal functions by tweaking it on occasion, are you against the tweaking on principle or how do you feel about it?

DR. COSGROVE:  Well, yeah.  So now we're treading on this tricky moral ground.  In my opinion our concept is to restore normal function.  In the same way that I would have an aversion to trying to use deep brain stimulation for social, political, or, you know, legal issues, I have also a problem with trying to enhance function artificially because it seems beyond what is normal for that person.

So I think of these interventions as trying to relieve suffering.  I think we can all agree that that's typically a good thing and a laudable goal, and then to try and restore function back to normal for that not beyond because then I think you're treading on very difficult areas.

DR. KRAUTHAMMER:  And just to follow up, do you think that's scientifically plausible?  Aside from the moral issues of whether you ought to do it or not, do you think it will be doable in our lifetime?

DR. COSGROVE:  Could you improve function?


DR. COSGROVE:  I believe it will be, and that's why I think it's important to have these discussions now when it's not an option at this point, but that there's a firm ethical framework upon which we will judge and make decisions going forward because I do think we will be able to improve certain functions in different individuals.

DR. HURLBUT:  What kind of functions?

DR. COSGROVE:  Well, this is just an example of creativity that was manifest in a single patient.  So if you could work harder and better and faster and do better work, then is that something — well, that's what we try and encourage our kids to do.

DR. HURLBUT:  What do you mention?

DR. COSGROVE:  Well, one could be, again, more creative from a scientific perspective.  You may have clarity of thought.  You might have better as we've said artistic endeavors.  I mean these are all scary things.  You might be able to analyze a situation more clearly.

Whether that's a specific effect or whether it's a relief of some underlying problem.  For example, on many of the patients in whom you improve with depression, their performance on neuropsychological tests often increases, improves after even appropriate medication or even a surgical intervention, and it's not because they are any really smarter or better.  It's that their mood is improved, which allows them to pursue and attend and function at a higher level.

So all of these things are very speculative, and I'm not typically a speculating type.

CHAIRMAN KASS:  Peter, Janet, and Gil.

DR. LAWLER:  So in your opinion, to summarize, right now this is for symptom relief in very extreme cases.

DR. COSGROVE:  Correct.

DR. LAWLER:  But it could be a lot more than that, but you're against that.

DR. COSGROVE:  Yes, at this point in time.

DR. LAWLER:  What if I had been a great neurosurgeon, but I was suffering from mild depression and I could save many lives if this option were open to me?  What would be wrong with that?

DR. COSGROVE:  Because there are other, much better alternatives for the treatment pharmacologically, and Ben probably is a little depressed, and he still saves a lot of lives.


DR. COSGROVE:  We all get a little depressed when you do neurosurgery for this long.

So I guess it all comes down to — it all comes down to the risk-benefit analysis, and you know, surgery always has risks associated with it.

DR. LAWLER:  I agree there's no ethical question if it's symptom relief in genuinely extreme cases, but what's to keep your colleagues from disagreeing with you as these procedures become easier and the cost-benefit analysis starts to shift on you?

DR. COSGROVE:  Well, when we have the evidence to demonstrate that we really understand what the outcomes are, what the benefits are, and what the risks are, then, of course, one reassesses the situation, and in the same evolution in epilepsy surgery it used to only be done for patients who had failed all anti-convulsive medications. 

In the past decade or so, we have evolved into the patients with medial temporal sclerosis as that's a surgically curable epilepsy, and we know that about 80 percent of those people with appropriate operation will be cured.

So we've evolved from only doing it for the intractable patients that failed medications to saying there is a subset that really looks good and, you know, we know what the outcomes are.  Therefore, the risk-benefit is good.

Similarly, with Parkinson's disease, we used to only do it for the end stage.  Now we do it for younger patients because we know what the track record is and we feel that there is an advantage to doing it when you're younger to maintain function for longer.

When we get the information  about deep brain stimulation for psychiatric illness that we can hang our hat on and know, then we might —

DR. LAWLER:  Well, now you're scaring me.  Then Leon's question kicks back in, right?  This is a radical stand.  I'm for complete eradication of Parkinson's disease.  I know it's controversial, but the same thing with epilepsy and so forth and so on.  But what you're saying is over time what seems extreme becomes less extreme because it becomes easier to do.

DR. COSGROVE:  No, not because it becomes easier to do.  It's easy to do this now.  That's not the difficult part.  I mean, it's easy for any good neurosurgeon to do this now.  That's the dangerous part.  It's easy to do right now.

What we don't have is the knowledge and the experience with it, and that is what's missing.

DR. LAWLER:  So it becomes more easy and reliable and comprehensible and stuff, right?

DR. COSGROVE:  It's much more predictable.

DR. LAWLER:  Predictable.  That's the word I'm looking for.

DR. COSGROVE:  "Predictable" is what you're looking for.  When we can predict exactly what the outcomes are going to be, then it makes it easier to make a decision.

DR. LAWLER:  Then the analysis for the decision is easier.

CHAIRMAN KASS:  And now much a technical point in follow-up, Janet, if I might intrude myself in the queue.  On the one hand, you seem to be saying that it's much too early to tell, but it's at least conceivable that once the data were in, that psychosurgery, precise psychosurgery of the sort we're now talking about and precise stimulation, might become not a treatment of last resort, but maybe even for certain kinds of conditions a treatment of first resort.

I mean if the trouble with obsessive compulsive disorder is that these people who have had it for such a long time are now hard to restore to work because, as you indicated, disease is so severe; there has been so much damage as a result.  Then one could make an argument if one knew what one was doing that you shouldn't allow them to have 20 years of this disease and that it would be more efficacious to go in there early, assuming you could —

DR. COSGROVE:  Predict outcome.

CHAIRMAN KASS:  Predict outcome.

DR. COSGROVE:  In that specific patient.

CHAIRMAN KASS:  In that specific case.

DR. COSGROVE:  So, yes, we have that same analogy in epilepsy.  We prefer to operate now on children with epilepsy because we know what the natural history is of that disease.


DR. COSGROVE:  To try and, you know, allow them to have a 40-year history of epilepsy and then operating on them when they're 40 seems a waste of time, but we have intimate knowledge of the natural history.  We have intimate knowledge of what the expected outcome will be from an intervention, and we can make it even individual specific because of our characterization, our knowledge of the disease, our understanding of the pathophysiologic mechanism.

So while we can draw an analogy to that, I think that, you know, I would not consider it as a first line of resort ever because there are pharmacological and behavioral therapies that are really the mainstay of treatment.

And as I've told you before, this surgery, ablative surgery or this surgery, does not cure these patients.  It improves sometimes just their response to behavioral therapy or their response to the medicines.  So it would not be considered in the same way we talk about surgical cures.  It's a different animal.


DR. ROWLEY:  Well, actually the question I had has been in part answered by these conversations.  I was reminded of my colleagues in oncology who started out with new drug therapies on the patients who have failed everything, partly to get experience with a particular drug or treatment, the dose, the scheduling, and all of the rest of it, and for those that appear to be efficacious, and then moving on to using them in earlier patients.

And so this conversation has said that, in fact, that might be a possible scenario as you gain more experience both with the type of patient that could respond to the treatment as well as all of the down sides of this treatment so that you're more comfortable going earlier in the patient's disease.

What I take away from this conversation in the last few minutes is that, in fact, this would be an appropriate, reasonable way to move, but you do indicate that surgery is not the first line for virtually any patient, but that there may be a subset of patients in whom some kind of surgical or deep brain stimulation intervention would be appropriate much earlier in the disease than you presently feel comfortable with, but you need more experience to determine that at that point.

CHAIRMAN KASS:  That's true.


PROF. MEILAENDER:  Just a question.  When you were discussing the surgical procedures and the percentages of people who are responders and partial responders and so forth, you said that the improvement is greater over time.  It's not immediate.  Why is that?

DR. COSGROVE:  That's one of the enduring questions, isn't it?  It speaks to our relative lack of understanding of the neurobiological basis of these illnesses.  We —

PROF. MEILAENDER:  Could I just sharpen it?  Is it certain — does it raise questions about whether the basis is entirely neurobiological?

DR. COSGROVE:  No, not in my mind.  I believe, you know, in the same way that it takes time for an antidepressant medication, you know, you have to have it in the patient for two to four weeks at appropriate levels before you begin to see any change in the neurochemistry and, you know, before you begin to see change in symptoms.  The making a lesion or stimulating the brain is sort of an internal neural modulation of some sort, and it's probably because, you know, it's not a specific nucleus that drives all of these things; that it's some rebalancing act that is occurring.

We don't understand why it takes that long and why we tend to see improvement over time, and you know, even some of the speculative mechanisms that we've used to explain some of these things are just wrong.  You know, we find out 20 years down the road that, well, it seemed like a good explanation for it, and it seemed to fit our empiric observations, but it was just totally wrong.

So I don't believe in speculation.  I have too much work to do to spend time speculating, and I am a fairly strong believer in empiricism if it's valid, but nobody really understands why it takes time.  I think it has to do with interfering with circuitry that allows the brain to remodel.


DR. CARSON:  Yeah, Rees.  That was a very, very informative talk.  I think it's going to give us a lot of food for thought.

One of the things that I think people need to understand is that neurosurgeons tend to be extraordinarily conservative people, and you can see from Dr. Cosgrove's presentation that he fits into that category.

I have no doubt that as time goes on and as people become more and more familiar with the techniques and as more less invasive types of techniques become available, that the degree of conservatism will slacken and that the number of applications will begin to expand, and this will become a significant issue.

You look at things like intervention at the level of the hypothalamus.  You know, if you ablate the ventral medial hypothalamic nuclei, all satiety goes away.  A rat will eat until, you know, it explodes. 

The same thing would happen to a person.  You stimulate the lateral hypothalamic nuclei, they're not going to want to eat.  Now, don't believe for one moment that somebody isn't going to try to exploit that, you know, when we come to all of this dieting and ways of getting people to be looking, you know, the way they think they should be.  All of these things are going to happen because we can do it.  We have the ability to do it.

Like Clinton said, "I did it because I could."  You know, it's going to happen, and we're going to need to deal with it.

CHAIRMAN KASS:  Could I follow that?  Because the measuredness and sobriety of the presentation I very much appreciate.  In fact, that's one of the reasons why you were recommended to us, and you didn't disappoint.

But you indicated that, on the one hand, there is a kind of unfortunate restraint based upon the past history which prevents these procedures from being reimbursed under Medicare, right?  This is by name specified as a no-no, and that now stands in the way of actually people doing these procedures as much as they would be indicated to actually help the people who were in desperate straits.

On the other hand, you say that whenever you get the neurosurgeons together, you ask them, "How many of you guys are ready to do deep brain stimulation for psychiatric indications?" and all of the hands go up, which means that there is an enthusiasm for doing this already, even in the absence of the kind of knowledge that we have, and if that financial constraint, if the reimbursement constraint were removed and you have also this kind of monopoly, you know, of the device driven and monopolistic character of the equipment, should we not be concerned that there is something ready to take off here in a few years?

I mean, or is this just science fiction and that we should rely on the good, conservative sense of the Carsons and the Cosgroves?

DR. COSGROVE:  No, you shouldn't because neurosurgeons actually — I'm not sure I agree with Ben entirely that they are a conservative group.  Neurosurgeons often operate first and ask the questions later, and in this regard neurosurgeons are not well equipped to assess the accuracy of diagnosis, the adequacy of treatment.   I mean all of those things.

They are equipped to do the surgery, but sometimes it's frightening to me how even ill equipped some of the people are that undertake the surgery, where I get phone calls about, well, you know, "I'd like to do a cingulotomy for such-and-such," you know, and they're calling about the coordinates, about how you do this.

And I have repeated, you know, experiences this way, and then at the end, after I'm thinking, well, are you sure you want to do this or are you sure you have support of your institution, because of the lay people's impressions and, you know, all of the trouble you can get yourself into.

And then, you know, at the end of the conversation where I've described everything and sort of put up all of the warning signals and then at the end, you know, the neurosurgeon says, "Well, are you supposed to do that on both sides?"


DR. COSGROVE:  And I think, "Oh, my God," you know.  So I'm not so sure that we're all going to be responsible practitioners.  You see, I'm very sensitized to the irresponsible practitioners of the past, and I do believe that this surgery, whether you use ablation or deep brain stimulation, is an important intervention.  Currently it's an intervention of last resort, and it is a good palliative procedure.

I do not think we will have another opportunity to do this.  So if we do not do this right and carefully and, you know, properly, I don't think it will come back.

DR. ROWLEY:  Can I just ask you a question?  Is there a role for the IRB here?  I mean, you would think that somebody couldn't just go and do an operation willy-nilly particularly in what is an experimental operation.

So where are the institutional safeguards?

DR. COSGROVE:  So one of the things, as I said, if we're going to move forward with this, it should only be performed by expert, experienced, multi-disciplinary groups, number one, with all of the people that I've proposed.

Two, it has to have institutional review or board approval for the institution.  And so, you know, it has to because it is an investigational, any intervention.

One good thing to say is that the company, Medtronic, which creates all of this equipment, is actually very responsible in its behavior.  It is not out there trying to get people to do these procedures.

In fact, they, you know, don't want these done outside of the context of a trial because they are also concerned about misuse and abuse, and again, if it's not done properly, well, their bottom line will suffer, and that's why they're interested in it.

But you know, we as practitioners and as leaders of society, we have to concern ourselves with not their bottom line, but the society's best interest and then the individual patient's best interest within that society.

CHAIRMAN KASS:  Bill.  We're coming to the end of this session.

DR. HURLBUT:  One very quick little question.  When you proceed from the costs, the practical and the aesthetic constraints on this, wouldn't this really be better than pharmacologic treatment?  It's more targeted.  It's more specific.  You're actually addressing a local problem instead of the global, systemic delivery of a drug.

And secondary to that question is are these techniques eventually going to be very valuable in explorations as well?  I mean, you wouldn't explore in a well patient, but we'll learn a lot from patients on this.

Can you combine this with local drug delivery through these same devices and can you do a micro electrode analysis of what's going on in the local area while you do it?

DR. COSGROVE:  Those are fabulous questions.

Yes, it is conceivable that this would be more locally specific than a medication taken systemically, although, again, it reveals our basic lack of understanding of the neurobiological basis of these illnesses, but we do know there's a lot of evidence that implicates the frontal orbital cortex, the ventral striatum, the anterior cingulate.  All of these areas that we've targeted in the past and that we're stimulating now have broad ramifications primarily in the frontal lobes, and cortical thalamic connections and striatal connections.

So, yes, it is possible that you could get a better, more selective, therapeutic effect with deep brain stimulation.  It's possible.  No evidence to suggest that that's so at this point in time.

Can you learn a lot about human behavior with these?  Yes, by implication you can stimulate different contacts, and can you by stimulating alter behavior and, therefore implicate those areas?  Yes.  You have to remember though that just because you're stimulating in one little area doesn't mean you're not stimulating afferents and efferents that are going to far reaching neural systems. 

And so, you know, it's not as simple as, well, if we push this button, if we stimulate here, that that is the seat of that function.

What was the other one?  Oh, could you inject through the catheter stimulator?  Yes, it's quite possible.  I mean, we do those kinds of experiments and injections in Parkinson's patients in whom we're studying.  We have an opportunity to study neural function at the target zone because we are there.  We actually use micro electrode recordings to fine tune our targeting.  While we're doing that, it's a fabulous opportunity to study human neurophysiology, and we actually undertake those experiments.

And when you're dealing with psychiatric and behavioral issues, you know, you can train a monkey to move a joy stick and you can time and you can analyze the motor systems much more easily, and we know much more about motor systems because we have monkeys that can do that.  We can train them, and we can train animals to do certain things.

But you can't train a non-human primate to make moral decisions about a — well, it's much harder to train them about reward and negative consequences and various things, whereas a human gets it like that.

So the ability to explore human brain function that is unique to human beings, yes, it does provide an opportunity.  I will say it's not easy to do that, you know, in an operating room with a patient.  You only have a short period of time and the set-up and the rig and all of the requirements to do that well is difficult.  You can certainly do it poorly and come up with all sorts of speculative reasons why this works, but, yes, in my mind it's a unique opportunity to understand brain function both on a macro level and a micro anatomical level.

CHAIRMAN KASS:  I'm going to wind up.  Indulge me one last question because you've sat through  the somewhat chaotic discussion in the last session, but could I bring you to comment?

If it were the case that neural imaging were able to give some fairly clear correlations now, without understanding causation, to identify populations of patients who have  difficulty controlling rage, and that borders on violence and aggresivity that you spoke about earlier, do you see a possible future that these things might not be controllable pharmacologically?  Is there a possibility that brain stimulation or ablation might be able to lend a hand here?

I know this is not what you get paid to do.  I mean, you get paid to deal with the people in the retail business, but as you were listening to that conversation, did it sound to you like something that might sooner or later come your way as a result of what brain imaging is going to disclose?

DR. COSGROVE:  So can I give it a little preamble about braining imaging and neuropsychiatric illness?


DR. COSGROVE:  So it's already there.  Helen Mayberg has shown that with PET studies you can predict almost with 100 percent accuracy which patients are going to respond to Prozac, one of the first SSRIs, by a demonstration in the anterior, most rostral cingulate gyrus of metabolic changes there.  If a patient has those changes, then you know that they're going to respond to the Prozac.  If they don't have the changes, they're not going to respond to the Prozac.

So there are already several other examples of the same thing.  So functional neuroimaging can predict outcome to drugs.  We have only pilot data both in our depressed patients and in our OCD patients.  We would love to be able to predict which of our patients are going to be responders because if we can get all of those nonresponders out of our pool, then all of a sudden our statistics look great and everybody would say it's a great operation, right?

We do the same thing in epilepsy, of course.  We do PET scans to try and ascertain, you know, the PET scan showing temporal hypometabolism predict outcome.  It's a much better predictor of outcome.

We have done this work in about a dozen patients in both populations, trying to correlate preoperative PET scans with outcome from surgery, and the remarkable thing is that in both the OCD population and the depressed population, there's one area that predicts outcome, and it's linearally correlated with improvement.  In the OCD population it happens to be posterior cingulate, well behind where we do our lesion.  That's the only area.  It has broad connections in that area to some of the areas that we're talking about, and in the OCD population it happens to be the right thalamus and the right orbital frontal cortex.

So these predict and, again, in a linear fashion.  So if this holds true, and we're trying to substantiate that with larger numbers, and this might be just an epiphenomenon.  You can never be sure, but it makes sense with our a priori knowledge of what systems are involved.  If this holds true, then we have a much better predictor of response to our intervention.

Now, getting back to your more thorny question of if, you know, somebody was presented to me with aggressive behavior and you know we had a predictor of response based upon neuroimaging, well, it would have to be a convincing predictor of response.

Then you would have to correlate that with your response to outcome or your outcome in response to the treatment, and then it would also still have to satisfy in my mind the two preeminent criteria that the patient understands the risks and benefits as we know them, and so, you know, assuming you have great data on that, and that they wish to pursue this and they're able to give their own informed consent, not that the institutional advisor in the penitentiary, you know, says this guys is bad and you should do it because then I think that you're revisiting the issues that were addressed in the '60s of mind control, violence in the brain, all of those thorny issues which did nothing to help the discussion and debate on the subject matter at hand.

CHAIRMAN KASS:  Thank you very much, and thank you really for a wonderful presentation —

DR. COSGROVE:  Thank you.

CHAIRMAN KASS:  — and very thoughtful stuff for discussion.



  - The President's Council on Bioethics -  
Home Site Map Disclaimers Privacy Notice Accessibility NBAC HHS