“Wired for Depression” with Harold Sackeim — Part 1: Introduction to ECT
Marom Bikson
Alright Harold let's start.
Chapter 1.1 First Exposure to Brain Stimulation and Research on Mood Laterality
Harold Sackeim
Well, I suppose starting at the beginning would be a good place. And the first experience I had with brain stimulation, believe it or not, was as a undergraduate graduate student at Oxford. I had completed an undergraduate degree at Columbia and received a fellowship to Oxford where I did a second BA.
And I had a lab that my first year there. which was taught by Edmund Rawls, who became one of the greats in the neurobiology of emotion and the physiology behind that. the first time I did brain simulation, had any exposure to it was during a lab taught by Edmund Rawls in 1972, the fall of 1972 at Oxford. And I was taught where Bregma was to find essentially the vertex on the skull of a rat and to put an electrode into the pleasure centers.
And to have the rat self-stimulate, learn to self-simulate. I don't remember whether the experiment was successful or not. What I remember was being blown away by the idea that animals would work relentlessly around the clock to produce an electrical signal that seemed to provide them with pleasure. It was addictive. And it was as concrete as you could get in terms of illustrating how electrical simulation could change affects.
Going a step further, I did my PhD after Oxford at the University of Pennsylvania. And I was recruited there by Marty Seligman, the father of learned helplessness and positive psychology. But I chose to work with the Gurs - Ruben Gur, who is a first year assistant professor. And the reason was Ruben was the only member of the faculty that said I could do the research I wanted to do, which I had been working on the concept of self-deception and do people lie to themselves? Could I produce a laboratory illustration of this phenomenon? Could I produce it in the lab and essentially prove that the body knew one thing, but the person was saying the opposite? That aside, that became in psychology a classic experiment and so on. But at the same time,
Reuben and Raquel were studying laterality and were just developing the very first brain imaging techniques, working with inhalation to measure regional cerebral blood flow in the brain and positron emission tomography to essentially measure CBF and metabolic rate for glucose. And while working with them, I became intrigued with some laterality issues in particular was a brain specialized left and right in subserving emotion. And I produced a paper that got published in 78 in Science, which was the first demonstration that the human face was asymmetric in its emotional displays.
And that's now been replicated over hundreds of times in many, societies. It's a fact about us. And at the same time, I started work with Ruben Gur and Norm Geschwind - Geschwind being the father of behavioral neurology on where lesions were in the brain in cases of pathological laughing and crying. So, you could have disorders where damage to the brain would produce these uncontrollable outbursts of laughing and or crying. Where were they? And as it turned out, you could also have those same phenomena occur as a prodrome to seizures. So, you have something called gelastic epilepsy, where people who go on to have a frank seizure will be laughing as part of their prodrome or less commonly, the dacrystic epilepsy.
Well, they'll be crying. And we published that paper in the archives in neurology, as well as a sample of people who had half of their brains removed, at least at the cortical level, hemispherectomy and what the emotional consequences of that were. work also suggested that lateralization played a role in the regulation of emotional expression, positive and negative, that laughing was more common with silent lesions ⁓ on the ⁓ right side and crying with silent lesions on the left side. But when it was epileptic, they reversed the laterality. So, when you're putting your foot on the gas pedal, in other words, you laughed when it was on the left. But when you were breaking the brake, the brake was on the right. So that produced a model that to this day, there are adherents to Davidson's work. Richie Davidson was a colleague and a friend who ran with that type of idea further in claiming very strongly that their affect is lateralized to left and right centers of the brain.
Chapter 1.2 Wired for Depression 1
Marom Bikson
All right, now I can see obviously there's a nice set up here to brain stimulation when it's gonna show up. But before that, why from an evolutionary perspective would sort of happiness and, or from a functional, why would happiness need to be on one side of the head and sadness need to be on the other side? Or is that not the right way to think about it?
Harold Sackeim
…anymore. I don't think anything is nearly as simple or discreet. the reason I basically and to this day, I believe we are wired to be depressed and we are wired to be for it. That I don't think we could love for instance and have attachments to others unless we are capable of experiencing loss, serious loss. And I also still hold the view that what's atypical or abnormal about the depressed state as a psychiatric condition is not the symptom manifestations nor the depth of the manifestations, it's the persistence that essentially depression is a case of bad breaks. And I think they have to be segregated because only in infancy do you see people laughing and crying simultaneously.
Bipolar disorder, you don't see the simultaneous phenomena. see, you can see rapid cycling where they switch from one state to the other. But these are not states that we simultaneously hold very often. And then in fact, there's probably a reciprocal organization that dampens one and ups the other. So, it's not a mistake that many antidepressants provoke mania, for instance.
It's and so that was sort of my thing. in terms of wiring level, you can begin to see the circuit diagram. I also think that. It is far easier in the human. To pharmacologically and through other manipulations, provoke euphoria. We have all of these euphorians around, whether it's opioids or cocaine or on and on. But we have very, very few drugs that will reliably produce depression. I think there's a difference in the extent to which these phenomena are under inhibitory excitatory control. On the other hand, it's the opposite that you can easily provoke psychologically depression.
I did work at one point in my career where I was in the laboratory, telling people that they did terrible things to imagine the worst event that ever happened to them and measuring lacrimal flow in men and women to these types of events. And it was much easier to get them to experience depression than happiness. So, I don't think that these are independent circuits. But the larger point was
In the process of doing this work, Ruben Gur and I had a book contract with McGraw-Hill where we were going to write a book that we never did. And we also never returned the advance. ⁓ It was a book called Brain, Thought and Emotion that was going to essentially review much of the laterality literature in terms of its importance in terms of thought. It's obvious in the language, for instance, but also motion, which was a emerging area.
Marom Bikson
…was around -what year was this?
Harold Sackeim
It was a 77, 79. I was a Penn from 74 to 77. And was, at the age of 25, was offered an assistant professorship at Columbia in psychology. So I was in the psychology, their only clinician, the only clinical psychologist in that department.
Chapter 1.3 First Exposure to ECT
And…I had Stan Schachter on one side of me and the chairman of the department on the other. You know, so it was a gust position for a very young person. And I looked in the ECT literature because ECT, number one, and I didn't fully appreciate it, certainly at the time, was the most effective treatment we had for severe depression.
It also was profoundly effective in mania. was so as a technique that very few do both. And this one did both. I just had a review paper in the American Journal of Psychiatry on 80 years of experience treating a mania with ECT. What's that about?
Marom Bikson
…by late 70s ECT is really well established.
Harold Sackeim
Well, by the late 70s, it was ancient because it was first used in electrically in 38 and seizure induction as a therapy was invented by Don Meduna in 35. First patients were treated in. So, ECT is like 90 years old in terms of its origins. It's.
Marom Bikson
When it first came into your sort of focus, late 70s.
Harold Sackeim
Right. It was the late. Yeah. And I had I had trained as a psychologist and I thought, you know, my gut sense was easy to use a terrible thing. How could you be doing this to people? The first date I had with the woman who became my wife, she asked me out was to see one flew over the cuckoo's nest. And … so there was almost a societal…if you had a bad guy in the field, easy to use it.
And, but as it turns out, and, this is where, as a scientist and also as a clinician, people really have to take a step back. This intervention was by far the most effective treatment we had for these illnesses. was absolutely life-saving. It, the patients who are being treated were the sickest of the sick with the illness. They were the people who didn't, who failed everything else that we had to offer in those days. And they got the most well and they did it without drugs. From as a scientific perspective, what a laboratory for studying the nature of emotion, what was changing in the brain that made this so so effective.
And from a brain stimulation point of view, it was the only intervention we had at the time that had spatial localization as part of it. There was a raging debate then, less so today, but certainly still there, as to where do you put electrodes on the head in order to produce this effect? And where do put them to reduce the cognitive effects? The thought and emotion were built into the ECT paradigm because you could do it bilaterally or unilaterally. So from my point of view, could, doing it on the left side, as good as doing it on the right side, let's say in treating depression, a very reasonable question because the physiology at the time, I saw this very early on, suggested that ECT was highly lateralized in its physiologic effects. The impact on evoked potentials, on EEG, were clearly lateralized. We did some of the very first work using modern imaging and showed that an hour after ECT, all of the changes you get in cerebral blood flow were on the side of stimulation.
Marom Bikson
To put it in context, saying at the time it was the only brain stimulation we had that was somewhat regionally targeted but were there any other brain stimulation at all? So, okay, it was the only brain stimulation in psychiatry and it was also an atomically targeted intervention in psychiatry.
Harold Sackeim
It was the only form of brain simulation period. So, for instance, if you go back to, let's say 30 years ago, 1995, there were about 250 papers a year on brain simulation in the world being published. All of them in on ECT. You know, everything else that we have came after. Right. And so that was an accident. The brain stimulation part of it - to the that ECT was of scientific interest and should be. You have something that works that well. And the cognitive effects were profound.
Marom Bikson
The negative cognitive effects.
Harold Sackeim
Yes, that at the time around 1980 or so, most of the patients in the world, virtually everybody was being treated with a sine wave stimulus at the maximum the device put out. And so the consequences of that is the average time it took somebody to know their name, the day of the week where they were averaged after a single treatment averaged about eight hours with bilateral ECT and six hours with unilateral ECT.
Marom Bikson
And just to clarify, so again, this would be not an optimized waveform. It's a sinusoid, like the kind of thing that would come out of a wall plug. And there was no titration, meaning no sense of adjusting dose or personalizing it. just, whatever the device provided is what it provided when you activated it. And so therefore, quite possibly people were being provided a lot more energy that was needed.
Now, was it so much that everyone was seizing? you didn't have this was still this was still seizure based, right? But it was so much that everyone had a seizure on the first stimuli. Yes.
Chapter 1.4 A Failed Study and the Origin of Dose Titrated ECT
Harold Sackeim
It's essentially that was true. And at the time I was funded. wrote a grant in 1979 to do an ECT study, which was a failed study, but it changed the whole field. The goal of the study originally was to figure out who needed bilateral, who needed unilateral to do measures of affective lateralization…to just like we do with language, where I had done some work for instance with commissurotomy patients in neurosurgery where we severed the connections between the two hemispheres and we inject sodium amytol into the internal carotid artery in order to barbiturate that side of the brain for a period.
And then we test language on both sides to figure out where that this was called the WADA procedure that Julian WADA invented as part of the troops in Japan after the Second World War. That's where this procedure actually came from. But it's still done to this day. You can see that having a method, because oftentimes, let's say doing surgery for epilepsy, if you leave the patient aphasic, that's a greater deficit than leaving them epileptic. Right. Right. So that you need to know this stuff.
And so, the proposition I had or the proposition I was presenting was…If we could measure where in the brain depression was, was it just right-sided or was it bilaterally represented and so on, then we could figure out which of these treatments we needed, you know, sort of simple concept. And that never worked out. You can't measure that stuff or I couldn't measure it never successfully in a way that would be at an individual level reliable, let alone is there really such strong lateralization? I'm doubtful. the way, just incidentally, this is what ultimately led TMS…this type of thinking to do the fast, left, slow, right. It's exactly that. all originated from that. You got to either boost the left, increase ⁓ activation there, or shut down the right with an inhibitory stimulus like one-hertz stimulation. That's all out of this original work with Gashawin and so on and then Davidson. And it's wrong.
TMS as a field never did the critical experiment to two by two because I believe that if you did slow left and fast right, the opposite of what the current mantra is, it would just work as well.
Marom Bikson
I mean, I want to get to the birth of TMS maybe as the child of ECT. We'll see. When you talk about this failed study, how are you planning to measure the lateralization? By doing ECT or by some imaging technique?
Harold Sackeim
We were using, the same techniques as we're being used with language, which worked well in language, dichotic listening to this is scopic viewing, as well as, you know, imaging results. But, but the, but, uh, beforehand, the, idea was to, know, before somebody got started, could you really tell what type of ECT they needed? And.
And that's where the non-invasive measures were developed. And I a lot of time up at Yale developing dichotic listening tapes with affective stimuli and, you know, that traditional psychology study that the Gazzaniga and others perfected. At the time,
Marom Bikson
It's failed in the sense that you're not seeing any sense, you're not seeing evidence that mood is lateralized, whether it's depression or happiness.
Harold Sackeim
I never developed a technique that I was confident you could measure it if it were lateralized. That it's a prior question. There may be some degree of lateralization to Mood. I'm not sure how important it is, but it was a failed study because we did something smart. And this was purely to do better science so that the art of the science was better.
I went to the experts in the field and asked, so how do you decide where to set a dial on this thing? Incidentally, this is a very different era. I was funded. I wrote the grant with Sid Melnitz, who was a psychiatrist at the Psychiatric Institute at Columbia, who needed somebody who was working in science and who had an interest in ECT. And Rochelle Gilman Klein, Don Klein's wife - great scientist on her own, Rochelle. She was the one that was the matchmaker between Sid and I. And I went around to the Max Fink and to Gabrons and the others in the field and asked them, how do you decide where to set the dial on this thing? They could not produce a rational answer. ⁓ there was no information on individual differences in seizure threshold as like you were asking, do people differ in their needs and so on? And out of the blue out, you know, I was trying
Marom Bikson
At that time there was bilateral and unilateral already?
Harold Sackeim
Unilateral was first used in England in 1956. There were many different unilateral placements. There was uncertainty as to where on the right side you would put the electrodes. And bilateral was always from the original Trileti and Bini in Italy placement because the thickness of the skull is less over these temples and many tongues.
That was the original justification for that position. Now there's also a bifrontal. And we know from much of the modeling work that where you put these electrons makes a huge difference, particularly with something like ECT.
“Wired for Depression” — Part 2: Emergence of Modern ECT
Chapter 2.1 Introduction to ECT and Study Design
Harold Sackeim (00:00)
So, in this grant that was funded, was written in 79 and was started in 1980. One of the aspects of designing the study was asking what would be an optimal way to stimulate the patient. And how could we compare two treatments when we didn't know whether we were dosing them appropriately?
Marom Bikson (00:28)
ECT for Electro-Convulsive-Therapy.
Harold Sackeim (00:30)
Right. And the field of ECT believed the seizure provided the necessary and sufficient conditions for efficacy. It didn't matter how you produce a seizure, whether you produced it chemically or electrically. And later in time, when we developed magnetic seizure therapy, you could do it magnetically. But as long as you produce the adequate generalized seizure, you had a potent therapeutic agent. And the biblical view is also that the electricity did not contribute to the efficacy of the treatment, but contributed to the side effects.
Right? So, but nobody in the world ever said, how does that translate into practice? Shouldn't I try to give the least amount of electricity possible, but make sure that the person has an adequate seizure. That way I guarantee efficacy and minimize side effects.
Marom Bikson (01:19)
To be clear, you've produced a seizure. The seizure is where the clinical benefit comes from, maybe some of the side effects. But as far as the direct effects of electricity, other than producing a seizure, at that time, those were assumed to just all be bad. So the electricity on itself is a bad thing. Only when you get a seizure at this point is something good possible. That was the thinking at the time.
Harold Sackeim (01:42)
And I realized this fundamental principle, which had some experimental support from Jan Adderwadison in Sweden. He had a brilliant set of experiments that were published in 1960, that, you know, 20 years before this work that ⁓ sort of set the stage. And so ⁓ the question was, well, how are you going to minimize dose?
If you go high and then go down, you have no idea where the threshold for the seizure is, how high above it you're starting. And it could take you forever to get to where you have essentially a subconvulsive stimulation.
Chapter 2.2 The Evolution of ECT Techniques
Marom Bikson (02:25)
Meaning if every patient starts with the dial to are turned all the way up.
Harold Sackeim (02:30)
And then you keep chopping it down. Because once you have a seizure, can't, you know, one of the things that we did know, by the way, is the seizure itself is anticonvulsant. It raises the threshold for itself. We could maybe talk about that in a bit. but the so we decided thinking through the problem to go below what would likely be the threshold. I was hoping to - maybe only 15 % of people have adequate seizure at the initial starting dose, a small minority with a very small dose. And because these steps are extremely large, we were essentially doubling the electrical charge at each step in order to try and capture this range.
Chapter 2.3 Titration and Seizure Threshold
Marom Bikson (03:19)
So, is this sinusoid or pulsed? Is this voltage controlled or current controlled at this point?
Harold Sackeim (03:25)
At this point, it's a brief pulse. It's with a Mecta device. I think it was a Model B. I have a picture of it on my office here. they actually, the guy who had designed this device was a brilliant, brilliant psychiatrist who had a premature and rather unfortunate death. Paul Blachley, who started the MECTA company, was at University of Oregon.
And he was the one that created the first brief pulse device to move away from sine wave, recognizing how inefficient the sine wave was in producing. And he also offered parameters in his very first device that disappeared for decades from the field. He offered a device that produced a 0.15 millisecond pulse width, ultra ultra brief. It wasn't until
2008 when we published our ultra brief work and moved the field down to a point three of a millisecond pulse width. And now we've gone back, we just finished a study randomizing to a point 15. So, know, 50 years later, we're returning to what this guy's inspiration was. He was rafting in the Columbia River with his son when his son fell when the river he jumped in and saved his son's life, but he drowned as a young man. And the people who now own what is Sigma Stim and previously the Mecta Corporation - at auction bought that company after Blachley's death. Interesting. Yeah.
Marom Bikson (05:00)
So, you're titrating, are you titrating intensity or your titrating _ _ okay so there's charge but you can, are you titrating the intensity of the pulses or you titrating the number of pulses?
Harold Sackeim (05:11)
Excellent question. And we'll come back to that. In developing all of this, we were limited by what the device could do at the time. The device was totally fixed in terms of its current. So it was an 800 milliamp pulse amplitude. Nothing you could do about it. Nowadays, I argue that the best titration would be in the current domain as opposed to number of pulses.
But I had the MECTA Corporation design for us a device. And the parameter that they could change the most and give us more space was frequency. So, we had a device from the very beginning that went from 20 hertz to 140 hertz. We doubled essentially the frequency range that it was capable of.
We let - my group later went on to show that if you're changing the number of pulses, it's more efficient to change the duration of the pulse trend and the frequency of pulses. so my devices when they working algorithmically always increased duration before we didn't know that at the time. The very first patient we treated was a woman about 35 years old, a schoolteacher, mother of two - who had been hospitalized at the psychiatric institute at Columbia for nine months. she believes she was burning in hell. This was her reality. Before she was referred to us for ECT, she put a rope around her neck and over the old pipes in the old building, tried to hang herself.
She was randomized. This was a trial in which the trial design was two cells, right unilateral, bilateral, for the very first time, giving the least amount of electricity possible at each and every session. This was a trial where we titrated radically. So if we found your seizure threshold on first session, we gave that same dose in the second. And in the third session, if you seized at that dose in the second session, we deliberately went down again - track seizure threshold throughout people's courses. that is the very one of the things that came out of that is in ECT, the threshold for seizure goes up about 60 to 70%, particularly people who do well. 60 to 70 % more electricity produces seizure at the end of the course than at the beginning. That's one of the very clear cut indications that it's anticonvulsive. This woman was randomized to low dose bilaterally ECT required four treatments and was completely euthymic. week later, she was back teaching school.
And I was impressed.
Marom Bikson (08:06)
And…that was to your first ECT.
Harold Sackeim (08:11)
Well, I had never seen the procedure before being funded for it.
Marom Bikson (08:14)
OK.
Chapter 2.4 Clinical Trials and ECT Efficacy
Harold Sackeim (08:16)
Would not happen nowadays. In those days, they sent site visit team out if they thought the grant was interesting. The site visitors came and we talked for an afternoon. They said, Yeah, we'll try it. And okay. And yeah, so that was my first experience with it. And what that study showed, it was published in 1987 in the American Journal of Psychiatry, the clinical results and the seizure threshold findings were published in what's now JAMA Psychiatry, also in 87, same year. And clinically what happened was right unilateral ECT didn't work at all. It had about a 17 % response rate. And whereas the bilateral group had a 70 % response rate.
Now we had managed undo the efficacy of right _ unilateral ECT. And that violated the fundamental principle of ECT because we had the most conservative criteria for what constituted an adequate seizure. And if it were just a seizure, was necessary for the efficacy of the treatment, we had managed to produce generalized seizures that didn't work. And that upset the field. That was not just you know, sort of this is a better way or not a better way of doing it, but sort of our understanding is wrong.
Marom Bikson (09:46)
Obviously, that wasn't the death of right unilateral?
Harold Sackeim (09:51)
Well, you could go two ways with that. Don Klein, my colleague, who is the head of therapeutics at Columbia and the at the time, the vice chair for research. Don is the father of psychopharmacology, literally. That's his. He said, Harold, you must have had a bad randomization. There's no way you could give ECT and have outcomes like that in a group. So that was one explanation is that. It was bad randomization.
The other was that dosage mattered for this treatment. What was a big thing that we changed? We changed, we gave bilateral and unilateral just at the seizure threshold. Okay. And so just in the second study in that series, we replicated those two conditions. And in every of the five ECT studies, randomized studies that we did at Columbia, independent of the multi-sector studies we did,
We always carried over two conditions, but in the next study, there were four groups. there were the unilateral and bilateral at just that seizure threshold. And then the same conditions where we turned the dial up two and a half times. And this is sort of to me was one of the beauties of working in brainstem and in sort of the biological basis of psychiatry is that.
You could do a really clean experiment. You can manipulate things in a very, very direct fashion. that study was published in New England Journal of Medicine in 93. And what that study did is it totally replicated the results of the first study that right unilateral at threshold doesn't work at all. Bilateral at threshold worked the same way, but took longer than bilateral ECT two and a half times above threshold.
Chapter 2.5 Innovations in ECT and Brain Stimulation
Harold Sackeim (11:42)
You got an acceleration of the antidepressant effect when you increase the dose. But with unilateral, you turn many, many more people into responders and remitters. And what it turned out was we had underestimated two and a half times I thought was above the level where seizure threshold is going up during the course of treatment that we would stay well above that so that we wouldn't have that problem.
But it wasn't enough. In the next study, which was published in what's now JAMA Psychiatry, there were also four groups. The two and a half time threshold groups were retained. And then we did six times unilateral and one and a half times unilateral. And basically, what we found is that at six times unilateral, it's as efficacious or more efficacious than any other ⁓ of the groups that there is a true dose response effect with unilateral ECT that when you turn the dial up, you get better and better clinical outcomes until you have to.
Marom Bikson (12:54)
So, there is a lot happening here in quick succession that ends up impacting the field for decades and impacting the lives of countless amount of patients. And we just want to pause and talk about here, right? One is the introduction of, I mean, things that would be taken for granted now, right? But here they are - coming up in very quick succession in a very sort of one paper after another, right? I can just see this acceleration of dosing. we have those first of all, you know, the systematic consideration of bilateral versus right. The first introduction of titration, which is done with controlled pulses instead of sinusoidal pulses. And then the realization that even once you have that threshold, you need to be considerate of a multiple of it. Right. And that that multiple depends on this montage, right, whether you're bilateral or unilateral. So, all this is, which is now foundation, like canonical, right, to the field is emerging at this time, right? So, sets the tone until today.
So maybe say more about that, but also I asked you before about how you titrated dose. So another thing that exists until today is that almost all or all ECT is done with a fixed current amplitude, which is not typical for any other space and brain stimulation, right? And this also seems to have, seems to be a consequence of decisions that you are making in this time period, right? That are just carrying through for decades.
Chapter 2.6 Challenges in ECT Research and Implementation
Harold Sackeim (14:30)
Yeah, I think the field of ECT to its credit certainly has moved forward and has accepted evidence. Maybe sometimes to its discredit, it's very slow in doing that. And some of that conservative in this comes from the criticality of the treatment and the disability of the patients - the impairments of the patient. So, you can imagine people are very reluctant to do change things that they think work. by that, and their, and their life saving. and so, one of the things we did at this time, to give you an example is we were the first to ever use ECT to stop status epilepticus. the stories about the first patients to ever have a deliberate seizure induced in them while they're seizing continuously. Incredible events and so on. That is now part of standard guidelines in the treatment of super refractory status epilepticus It's all over the place now. But it took, you know, 50, 40, 50 years for that to...
Marom Bikson (15:54)
That approach, well, counterintuitive then, now it became intuitive as a direct consequence of you first titrating threshold and then tracking how that threshold is changing over time and then realizing.
Harold Sackeim (16:09)
Right, and what we were able to also do is show that ⁓ seizures get shorter and weaker in their expression during the treatment if you hold dosage constant. The thresholds go up. If you look at glucose and metabolism and cerebral blood flow, you're getting marked reductions after treatment. Jerome Engel at the time out at UCLA, one of the greats in epilepsy, had published in Science the same sort of thing that with epileptic seizures you're seeing in the post-ictal state.
This inhibitory state, presumably tied to the processes that terminated the seizure itself. And so, we had the physiology, we had the concrete data on thresholds and so on. I spent two years following patients with epilepsy with the head of outpatient epilepsy at Columbia, because I felt it was wise to really get a better understanding of the effects of seizures on people by being not just with these ECT patients, but patients with epilepsy. And it was that connection where we saw patients together and talked about what was happening that led the epileptologist to say, yes, why don't we try this for the first time? It wasn't just my team's courage to do that. It was really the epileptologist's courage. Stan Reiser was the fellow.
Chapter 2.7 The Future of ECT and Current Titration
Harold Sackeim (17:39)
The first patient we treated like that was an 18 year old girl who had grand mal seizure since birth. She wore a helmet because she already had cerebral cicatrix in her CT at the time. This is before MRIs existed. And when we gave her ECT under oxygenation, anesthesia, you know, the full thing, three times a week, she became seizure free of spontaneous seizures.
And we did ABABAB and we showed that we got, after we stopped the ECT, we got epileptic rebound and then she went back to her baseline rate. And then we reinstituted ECT, all the seizures went away and we tracked with xenon 133 CBF, the changes in their cerebral blood flow in her brain with the ECT and her seizure frequency.
I presented that work in Montpellier in a satellite of the International Cerebral Blood Flow Metabolism meeting, an epilepsy satellite, and was almost run out of town. These were people that had devoted their lives to stopping seizures, usually in animal models. And here was somebody who was measuring seizure threshold in the human for the first time and deliberately giving people seizures and so on. So, there was a rather frosty reception initially - to that sort of thing.
Marom Bikson (19:07)
So, on this notion of not titrating to current level, right? And how maybe a bigger theme about when is brain stimulation an innovative field and when does it get locked into something that it gets into some pattern that it can't escape as a consequence of the machines that you were provided at the time, which at the time were innovative, but they did not have built-in ability to adjust the current threshold. You had no choice. You had to adjust thresholding through other means, frequency, number of pulses.
And boy did that stick. again, I want to emphasize it doesn't, I don't think that exists in any other field of brain stimulation or neuromodulation and all other fields. think adjusting current is a go-to, even a starting point. yet something that happened back then locked the field of ECT in for decades.
Harold Sackeim (19:59)
Yeah, I've written a fair amount about that and we just had a paper in brain simulation with Ziad Nahas where we randomized people to a 0.6 amp or 0.9 amp stimulus.
Chapter 2.8 Conclusions to Part 2 and Reflections on ECT
Harold Sackeim (20:16)
I believe that in the future, titration in the current domain would be a superior way to go. And I think it's a clear cut issue. Only a small portion of the electrical stimulus actually enters the brain. Most of it is shunted between the electrons, largely because the impedance of the scalp and brain are extremely low. Whereas the impedance of the skull is around 17,000 ohms per centimeter of depth.
And we're talking on the skin in the same units around 200 ohms per centimeter depth. So you're going to have a lot of the stimulus being shunted across the scalp, very little getting in. But how much gets in is going to be functioned largely with skull anatomy. And so there are two aspects here. What happens is people get different levels of current depending on their skull anatomy because that's what's being shunted is not the number of pulses, you don't do anything to the timing of pulses in the brain. What you do is you're changing the amplitude of the pulse when you're shunting. And
Marom Bikson (21:26)
Different people for the same stimulus current, different people will get a different amount of current into their brain as a result of differences in anatomy and tissue property.
Harold Sackeim (21:35)
A great example is females have a lower seizure threshold than males, let's say. much of the data, we reported that in 1987, much of the data still confirms that. But what do we do with males? We give them more pulses. So, the female is having a lower threshold because she has a thinner skull, right? More of the current is actually entering the brain. So, it's going to go to a deeper level in the female brain because of the, at least the depolarizations will because of the greater amplitude of the pulse. And to make up for that, we give a shallower stimulus with more pulses to the male. Makes no sense. And the problem clinically is, I also argue, we don't know where in the pulse amplitude range, we want to be in order to preserve the efficacy of the treatment.
It's not a, for instance, the current Sigma device has a current adjustment that you can go from 500 to 900 milliamps on it. That's a fairly substantial, but we don't know whether you should be at 200 milliamps or 1.2 amps. That what we're shooting for, what the average should be, what is an unknown in ECT, I argue.
And the president of a company, Somatics, who make the Thymatron, a paper in Germany, they did a study in Germany, where they double titrated people comparing a 900 milliamp stimulus to a 1.125, I believe it was, higher and claimed that you needed less charge to produce the seizure with the higher than the lower, that ECT devices are under-currented, so to speak, in terms of what would be optimal. There are people like Liz and others who are arguing that low amplitude is the way to go. And that's largely been motivated because you can't get much juice out of a magnetic seizure approach, that the amplitudes induced in brain are far below what you can achieve with electrical stimulation. And the hope is that if magnetic seizure therapy works, then we can mimic it electrically and we would go to a low amplitude approach.
Marom Bikson (24:12)
But of course, we don't, know, you're titrating based on charge, but ultimately, we don't care about charge. not like you're trying to save on your electrical bill. It's really about maximizing the efficacy and minimizing the side effects.
Harold Sackeim (24:25)
Exactly. So, to do that, you need to know what is the amplitude I need to shoot for. And around that, I can titrate. And we still don't have that information. In fact, the only study that's ever been done is the one that we did with Nahas that actually charge balanced, gave two stimuli to the same person - one at lower current but with more pulses, the other at higher current and fewer pulses and ask which is more efficient in inducing seizures, which results in a faster orientation recovery, which is a great acute marker of what long-term cognitive effects would be.
Marom Bikson (25:13)
But that and the other efforts you talked about, these are in the past years. There's still the fact that for decades.
Harold Sackeim (25:21)
Absolutely. The answer to that to this day, it's a big issue in the field. A lot of the field is now recognizing current would be the way to go for titrating, but nobody is really stepping back and saying, but around where am I going to titrate? It's just producing a seizure that is going to be consequential, which is going to be the optimal stimulus. So, yeah, that's certainly and it's evident that if you look at Vagus Nerve Stimulation, DBS, not so much TMS, but some of the other interventions that they're all adjusting current, essentially maximizing current and being limited by side effects.
Marom Bikson (26:04)
You try to find that therapeutic window first with current, then with other parameters. before we get back into the details, as a big picture point, ⁓ does I see it you're making a lot of the innovations? There's other contributors as well, but you're carrying a lot of them yourself.
Why in ECT? in other, is that a, you know, is this something peculiar about ECT where if you were to line up all the innovations that were made and there's maybe 10 of them, you eight of them are Sackeim papers. One person seems to be making so much of an impact. Is that true in other fields? Is that a general issue with medicine? Is there something with brain stimulation where it gets stuck and it takes potential disruptors to change things?
Harold Sackeim (26:56)
I think ECT is a peculiar place. Number one, my experimental work with ECT was done at the Psychiatric Institute. There are only two places in America where you could have done that work. The average length of stay of my patients was three to four months. These people were washed out of medication. They went through nine to five neuropsych testing before being treated, all sorts of physiological imaging work and so on. If they fail their first course of treatment, we automatically offer them a second course of ECT. And that's because at Psychiatric Institute, all the beds are free. If patients had insurance, we didn't accept it. This is like the NIH in terms of it being a true research institute with research patients where there's no financial aspects.
And at the time, you can imagine these are very, very sick people that require a lot of care. And when you're taking them off of medications, you're limiting their access to benzodiazepines and so on. You can't do it in a lot of places. So that was one thing that ⁓ because the state of New York - provided the support for the patient care where it was free. And also provided the support for a lot of the investigators, people like me and Joan Prudick and many others, the - Yvonne and Lisanby - That was extremely important in terms of our grants being competitive. The other part of it, I think, is political.
For instance, nothing like that ever came out of NIH and never would because ECT just elicits a political nightmare that leads to a dearth of funding. I tell my colleagues sometimes, very few people come home at night and say to their kids, hey, I'm a shock doctor. It carries its own level of disregard, at least in the funding process. And then anytime you do research with really, really sick people, where you're doing randomized research and so on.
It's really very, very hard. You have to have a lot of resources and support to do that. there's not very much ECT research going on at the level of RCTs high quality and so on. Much of what's reported is usually observational.
“Wired for Depression” — Part 3: Developments in ECT
Chapter 3.1 Reducing Cognitive Side-Effects of ECT
Marom Bikson
All right, so Harold, last time we left off, we sort of built up to what was really your first experience doing ECT and how much of the decisions you made at that time, including titration, set the stage for what would come for decades to come as far as ECT technology. So, I wanna pick up from there and really talk about what were the most important developments in ECT since you really started these innovations.
And also thinking a lot about both not just the efficacy, but the side effects, right, which are often discussed in the context of ECT.
Harold Sackeim
Thank you, Marom That efficacy of the treatment in large part is unchanged from the early days. When we were using very ineffective treatment and treating well above the seizure threshold, that did not compromise the efficacy of the treatment. What it did is it totally aggravated the cognitive side effects of the treatment. And as I indicated last time, one way of looking at it is that in the days in which sign waste stimulation was being used bilaterally, the average time it took somebody to know their name, the day of the week where they were was about eight hours following a single treatment.
Nowadays. That average with let's say ultra brief, right Unilateral ECT is seven to 10 minutes. Right? So, so this, and that change in the acute cognitive effects was mirrored in the more persistent and to the patients more concerning cognitive effects. to sort of put it in a nutshell, ECT in the 70s in particular was studied by the neuropsychologist extensively because it was a profound laboratory for the understanding of memory.
Chapter 3.2 ECT and (the study of) memory
It was an ECT that people like Larry Squire, who became president of the Society for Neuroscience and one of our great memory researchers, he came up with the distinctions between propositional and non-propositional memory, between episodic and non-episodic memory, declarative and non-declarative memory. Those are essentially all in the same types of comparisons, but he came up with them in part through the study of patients getting ECT.
And what was true in ECT was that when you came in for this treatment, you were intentionally impaired very often. Depressed patients, for instance, have often marked difficulty in taking information in about the world. Because, you know, if you're so focused on your ruminations and your depressive ideation and affect that it's hard to attend.
So, if you give the depressed patient a list of 10 items and ask them to tell it to you back, they'll report back immediately less than a person who isn't depressed. However, if you ask the depressed patient two hours later to tell you what was on that list of 10 items, the percentage they will report will be about the same as somebody who isn't depressed in terms of what got in. So, what gets in stays in before you get ECT.
ECT has the inverse effect on new learning and memory to the extent that it improves your attention as you get less depressed, but it interferes with retention of information over a delay. That's called anterograde amnesia.
Marom Bikson
Which is short term though….
Harold Sackeim
Yes, that was how do ECT… that last that deficit and learning and retaining new things last maybe two weeks, one week afterwards and so on. It's never really been demonstrated to last more than a few days. it clinically is very important because you know, you don't want to go out driving. You don't want to be making important decisions when you can't retain information over delayed disappears. But
That was one aspect of the memory deficit. And there, again, Squire's work was very important because Larry showed, for instance, that if you teach depressed patients mirror writing, sort of the equivalent of non-propositional, non-declarative memory, where it's like learning to ride a bike, there's no deficit after ECT. What they've learned stays in. And there are different neural systems that mediate these types of memory. And so you can see how important
ECT was in terms of how beginning to have an understanding about how memory is organized in the brain, what the essential features of it were. But clearly patients were reporting another type of amnesia, a loss of memory, which was more concerning to them and in many cases more persistent. And that was the loss of memory for events in the past. And many of the greatest critics of ECT were people that would say, you know, I lost five years of my life. It was just a blank. I don't have access to it. And it was very clear from a scientific point of view that in fact, if you presented information just before the ECT treatment and tested for its retention afterwards, that much of that information was vulnerable to loss. And for instance, if it was verbal information, the loss was greater with bilateral than with right unilateral ECT. But for spatial information, it was greater with right unilateral than left unilateral, that sort of thing. That was an important mediator.
Marom Bikson
But I can tell you a quick anecdote about that I had a family member, she was a young woman and ECT, she was extremely depressed and as with many people, ECT basically saved her life. And she was recounting a story where her mom would drive her to her treatments. They would be chatting in the car and she would get her treatment, she'd be walking home. And she was saying it was funny one time her mom was saying, she started to bring up a conversation.
And it was the same conversation that had on the drive over and she was telling her we already had this discussion. So, right, so transient and short term deficits are there. ⁓
Chapter 3.3 ECT Reluctance to Recognize Lasting Side Effects and Development of Technique to Measure Them
Harold Sackeim
But in the retrograde area, there clearly were people that were saying that ⁓ they were not transient, they were permanent. And in some cases, quite extreme, that years of their life was missing. And that became a highly politically contentious issue for ECT, in part because of the psychiatric profession, which, you know, certainly in the 70s and 80s and so on with things like Cuckoo's Nest and the anti-psychiatry movement, particularly the Church of Scientology, got very defensive about this treatment. And in the absence of objective evidence of persistent amnesia, essentially claimed that this type of side effect was not rare, but didn't occur, that all of the cognitive effects of ECT were transient.
And that was common in terms of what was being presented to patients. And particularly since my work was an ECT was funded by the National Institute of Health and our job was to sort of tell it like it is, we spent a great effort in terms of developing techniques that would objectively measure the loss of memory for events in one's own life.
And to give you a sense, in the first studies we did on this, we developed questionnaires that took about three hours to administer to depressed patients, asked about 300 questions about their lives, got 300 details about each patient's life, which we verified with a family member by asking them independently the same interview beforehand. And we did this with normal controls as well to see what the magnitudes of deficits were and so on. And these would be things like what was the address at your job? Not currently, but the address of your job before last. What was the telephone number? What did you do last New Year's Eve? Who were you with? What was the trip you took out of town, the trip before last? On and on and on. We had asked positively valence things. What was the best thing that happened to you in the past year? What was the worst thing? And to look at whether there was any valence to what memories were lost and so on. And we were particularly interested in whether there was any association with the patient's own experience of memory loss, what were the patients reporting themselves about memory?
Chapter 3.4: Without ECT Titration, Patients Are Overdosed
So put it in a nutshell. When we did the work introducing titration, what became obvious was that many patients in the old days were being treated, if they were using the same electrical wave form, let's say a brief pulse, many were being treated 50 times their threshold or more, 60, 40, 60, 50 times their threshold because there's great individual differences in seizure threshold.
That amount of electricity you minimum need to produce a seizure varies remarkably across people because everybody was getting the fixed dose at the top of the machine. Those were the thresholds were being those greatly over that immediately. Low dose treatment from the beginning was…was clearly less impactful on cognition in all respects, including the acute effects in that you saw in terms of orientation recovery or performance on neuropsychological tests right around the treatment for short-term effects, but it also turned out in long-term effects.
Marom Bikson
It's just an important point. just want to kind of clarify it, right? So, you could have one group of people that are just getting the maximum on the machine. They are getting, the ECT is still producing a seizure in them and they are getting the benefit. You have another group of people where everyone is being titrated. And so therefore no one's really at the top of the machine. They're getting less electricity. They're still getting a seizure. They're getting the same benefit, but less side effects.
And so, does that teach us that the side effects are being caused by the electricity and not by the seizure?
Chapter 3.5: Rapidly Shrinking Retrograde Amnesia
Harold Sackeim
Yes, there is no doubt that the nature of the electrical stimulation has a profound impact on the cognitive effects. May not be, you know, 100 percent, but it's certainly a great deal. It was all a package to the extent that the disorientation you see acutely following a seizure is to me a rapidly shrinking retrograde amnesia.
That things that come back first are the things you've learned earliest in life. And the things that most, that come back last are the things that you've just learned, like the day of the week or where you were and so on. And you can see in ECT, particularly with the old forms of treatment, that if you ask somebody how old they were immediately after they would wake up and they say they were 20 when they were 70 and they would age in front of you. They might be speaking their first language. And the psychoanalyst would call this regression, but the neuroscientists were saying this is a rapidly shrinking retrograde amnesia. And so, when you do that at each treatment and you don't have full recovery from a treatment, you could be producing a retrograde amnesia that's persistent and more extensive in its path.
So that for me was one of the reasons why it was wonderful that the period of disorientation was a biomarker for what your memory would be like six months later. And so, you could always clinically adjust the stimulation parameters with that in mind. And that's what we teach nowadays. But we were fortunate, the National Institute of Mental Health funded a study in which we looked at ECT outcomes at seven hospitals in the New York City region, not in...and not including Columbia, it was not a site. So, we had research assistants at these seven hospitals testing each patient before the ECT, during the treatment, and again, six months later. And that study, there was a paper in neuropsychopharmacology in 2007, which was called the cognitive consequences of ECT in community settings which essentially showed that how you were treated, where electrodes were put on the head, what the dosing patterns were, determined how persistent your cognitive side effects were. That for instance, bilateral ECT was associated with much more loss of autobiographical memory than unilateral ECT. that essentially in stimulating the brain, we had control over the magnitude of these events, not just immediately, months, months later.
What was also consequential, that was one of now a set of studies that showed if you ask patients originally, has ECT done bad things to your memory? As a group, patients say, no, it's done good things. And the extent to which they say it does good things, co-varied with your depression scores. That extent to which it said ECT did good things to patients, believe that was co-varied with their depression, because the more you had a clinical benefit. so ⁓ at one point we were actually in consent forms that were being recommended by the American Psychiatric Association for ECT said, however, a significant number of patients believe ECT has improved their memory.
Chapter 3.6: Believe ECT patients
Because that's essentially what they were saying. And they were saying things like this on the Squire memory complaint questionnaire and Broadbent's cognitive failures questionnaire. It turned out another contribution we made was to simplify things and say, let's just ask patients for their global impression. That ECT hurt or help your memory as opposed to asking much more atomistic questions like, do you have trouble reading road signs? can you concentrate on a book? Which is how the other question is when we asked a very simple question, does the helper hurt you on a Likert scale? That majority of patients said it hurt as opposed to help. And more consequentially, the other question here did not go very with any neuropsychological objective performance, whereas the global impression, the simple one, correlated strongly with patients' neuropsychological test scores but in particular most strongly with their degree of amnesia for autobiographical information.
Marom Bikson
…need to believe the patient, believe your patients when they tell you things.
Chapter 3.7: Reduce Side-Effects Promote ECT Adoptions
Harold Sackeim
Yeah, exactly. So, if anything, I mean, this is a wonderful story in the history of ECT because the magnitude of the side effects, which was one of the biggest rate limiting factors, ECT used to be in the very original days, over half the patients, their treatment was prematurely stopped because they were so confused. Their memory was so bad that that the treatment actually couldn't be conducted. They weren't recognizing family members necessarily. Nowadays, I've had patients where we've treated them and they've appeared on Broadway that night. And so, a whole different level of experience for the patient in terms of impact on cognition.
Chapter 3.8: Optimizing ECT Required Recognizing Side-Effects
Marom Bikson
You mentioned before that there was some, especially when you were starting to do ECT, there was maybe a reluctance among many practitioners to admit that there were these cognitive deficits. And it's almost because of your willingness to admit they were there and measure them that you sort of then had the opportunity to improve on the technique to make them go away, right? Those two things had to go in hand. You sort of had to be very willing to embrace the benefit and the risk in order to then be able to optimize those, to titrate those out.
Chapter 3.9: Brain Stimulation to Interfere (or Enhance) Cognitive Performance
Harold Sackeim
Right, I think absolutely right that why would you want to reduce, you know, the parameters of the treatment soften the electrical impact unless it was consequential and the area that clearly seemed to be consequential was in terms of memory. And for the field of brain simulation, this is one of the, to me, one of the really outstanding examples of the extent to which the parameters of stimulation dramatically impact on cognition. I mean, we're struggling in, as you know, clearly, Marom to identify manipulations and brain stimulation that will enhance cognitive performance. That's the goal. But we have this incredible track record of knowing how much we could interfere with it going the other way.
Chapter 3.10: ECT is Not Frozen: Ultra-Brief Pulse
Just one other comment on that - and ECT is certainly not frozen. I think it was in 2008, we reported the first major randomized trial of ultra-brief stimulation. And that was essentially, again, showing the same point with ultra-brief simulation, all we did was we took the pulse and we shortened its width. So, we compared all of our work before then had used a 1.5 millisecond pulse width. It was right in the range the devices were giving between one and two milliseconds. All right? And so we right in the middle and we were titrating with that. said, I thought if our understanding is that the chronaxie for getting a mammalian neuron to fire is between point one and point two of a millisecond. Why are we giving a pulse that's an order of magnitude wider? would we have lower thresholds for seizure by going to an ultra-brief stimulus? again, Lower charge, right? Overall lower charge.
And so, we randomized people to a study in which they got there were four conditions. Bilateral ECT two and a half times the seizure threshold, which is the highest you ever go with bilateral because the side effects become unacceptable above that, particularly with the brief pulse. And then right unilateral ECT at six times threshold because we'd already established with brief pulse that that's what you needed to make unilateral as effective as bilateral, but we didn't know for the ultra-brief what would happen there.
And the brief pulse was a 1.5 millisecond stimulus and the ultra-brief was a 0.3 millisecond stimulus, so a fivefold range. And what we found is that seizure thresholds on average were three to four fold less in terms of charge with the ultra-brief than the brief pulse. We didn't totally make up the fivefold difference, but almost.
And so, it was clear that the wider pulse was inefficient.
Marom Bikson
From a charge perspective, what about from a treatment or side effect perspective?
Harold Sackeim
If we take for instance time to achieve full orientation, it was a third that we got with it. that it would typically for the for the brief pulse bilateral, it was about 45 minutes. went down to 15 minutes. It went from 30 minutes for the unilateral six times threshold brief pulse down to about 10 minutes. it was a third the time it took. And that was totally reflected in how much memory loss there was for events in one's life. Essentially those orders of magnitude of difference. There are very large effect sizes between all of these things that were totally modulated by the nature of the electrical stimuli.
Harold Sackeim
So ultra-brief, for instance, regular lateral in many places in the world now is the default treatment. That's where people will start. And if it proves not to be work as well as one would like, then they have the options of switching to other forms of ECT. And to finish that story, we just finished a randomized trial in Prague where we compared a 0.15 millisecond pulse width, which we call ultra-ultra-brief, 0.3.
And people were crossed over at the first, they were titrated twice. And the point at 15 had lower charges associated with it and was more efficient in producing seizures again. So getting closer to the chronaxie for ECT is probably a wise idea.
Chapter 3.11 Spectacular Efficacy to Begin With
Marom Bikson
It's interesting, correct me if this is right. You're basically saying as far as efficacy, even the most antiquated form of ECT might've been as effective as what we have now. And so really the research on ECT in the modern era hasn't really focused on necessarily increasing potency or efficacy, but rather just reducing the side effects. Cause you were already starting with such a spectacularly effective treat, which is the thing unusual. I'm thinking most other fields of medicine.
Harold Sackeim
When we got to TMS, TMS had to grow in its efficacy, but ECT had reduced its side effects - we were blessed to have that opportunity to do that. But you can imagine, again, this is a context where the patients getting ECT at the Psychiatric Institute had an hour neuropsychological battery at every treatment, a week of neuropsych testing before starting ECT, then they repeated the whole thing afterwards. Then if they didn't get well, they got a second course of ECT. And then they were brought back at two months and at six months and so on. So, highly intensive in order to study all these things.
Marom Bikson
But the single most important thing that changed from, let's say, not modern to modern ECT was charge titration, which was actually, we talked about this in the last part of this interview, the exact place where you started. So, in fact, your very first innovation, your very first trial, adding in charge titration and then as we talked about recognizing it's actually different for different montages, that was essentially the most important breakthrough.
Harold Sackeim
That was...
Marom Bikson
That rest is icing on the cake.
Chapter 3.12: The Art of Doing a Better Experiment
Harold Sackeim
I would say yes, I would agree. that's my colleagues like DP Devon on head of geriatric psychiatry now at Columbia. That's essentially what he said was that the whole kettle of fish started with doing dose titration. That changed our understanding of ECT and how to give it and all that. Absolutely true. That it was all built off of that. And the thing, you know, for other people in science that I think is consequential is
That was not part of the grant. That was not anything we were ever really funded for. It was something that we thought would make for better science. That it was the art of doing the better experiment that said, I don't understand this, let's try it. Why don't we go about it this way? And it was the whole purpose of that first study was a failure, but...
It obviously had some consequences afterwards.
Marom Bikson
I agree with that, think, philosophically so much. We have this cartoon version of science that you stare at the ceiling, you develop a hypothesis, and then you just run it through. Where in reality, it's more about these observations you make in the process of doing something else, realize it, right? A failed experiment that instead leads to another discovery. Because it's almost like if we knew where to look, well, we would have looked there already, right? It's all about where you didn't know to look.
Chapter 3.13: Durability, Relapse, and Treatment Resistance
Harold Sackeim
I guess one last issue with ECT, which again, I think is very consequential for neuromodulation, is the question of durability. That's sort of the term that's being used most often now for this area. And again, when we started to work in ECT, the biblical view was that if you didn't do anything after ECT, 50 % of patients would relapse within six months. But if you gave an antidepressant drug, 20 % would relapse in six months. That was essentially the understanding. And what didn't make sense to me,
At the time, there was a couple of things. From the very beginning, we followed very rigorously all of our patients afterwards. And our social workers would call them up every couple of weeks and administer the Hamilton over the phone. And we would track relapse essentially for the people who had benefited from the treatment. And part of that was because of the feeling of the need to know, do you really go through all of this for...you know, a benefit that may last only a few weeks or is there some greater lasting benefit that you're hoping to achieve? if it is, how do you get to having that lasting benefit? One of the things I didn't understand was why we were getting so many patients who failed drugs before coming to ECT, who were being put back on the same drugs after ECT that didn't work beforehand.
And I also didn't understand why we were seeing so much relapse. And I remember going to one of the truly one of the greats in our field and asking them, you know, how come some of these drugs are not working for so many people? he turned to me and said, Harold, that's not an ECT problem, that's a drug problem. And our early work developed the first methods to quantify treatment resistance.
How potent was the anti-depressant regimens you were on or had received in the past? Which were they? How do you categorize them? All that stuff. We developed, one of the things I developed was the anti-depressant treatment history form, which is now the most widely used instrument in the world for evaluating the adequacy of anti-depressant treatments and so on. So, an example is in the current recovered trial of vagus nerve stimulation.
That patients in that trial average 13 failed antidepressant treatments historically. They're massively resistant. And so, one of very first things we reported back in 1990, that if you're treatment resistant, your likelihood of relapsing following ECT was twice that than if you had not failed an adequate medication trial before coming to ECT.
Marom Bikson
Where people getting ECT without having failed…
Harold Sackeim
Many people because some cannot tolerate an adequate medication trial. That was particularly true in psychotic depression where the dose of antipsychotic in need for it be considered adequate is often intolerable. Those are sort of the intricacies of the pharmacology there. Or people who were acutely suicidal had benefited from ECT in the past and nothing else had worked, those sorts of things.
So, we raised that issue, those treatment resistance predict, and now we've replicated that, we've replicated a number of times and that's now been replicated also in TMS and in other areas that treatment resistance predicts treatment resistance, both acutely and in the long run. Yeah, yeah.
Marom Bikson
…to anything, to drugs, you resist…if your treatment resistant, it's going to be harder and harder to get you to respond.
Harold Sackeim
For what it's worth, I'm controversial in believing that some of that is iatrogenic, that antidepressant treatment is like antibiotics. Antibiotic resistance is because of the exposure to failed antibiotic treatment. And I think that's also true to some extent with the biologic treatments, that they're not neutral in terms of their residue.
Chapter 3.14: Wired for Depression 2
Marom Bikson
I really want to go back to this question, right? Why then do you relapse? Hey, you're better, right? I mean, we see relapse with everything. It's not a uniquely ECT issue. So, did you have an answer to that issue?
Harold Sackeim
It's a complicated story because at the time, you know, back in the seventies and eighties, the biblical view in psychiatry was that depression was an episodic illness that would last for maybe two years on average, particularly severe depressions, but would spontaneously remit. That all our treatments were doing, and this I do agree with, was suppressing expression of the episode.
All right. But if you, but the underlying biologic disturbance continued despite that suppression. And if you took it away, the drugs before there was a spontaneous resolution, you would have relapse. And the early work in England on ECT was some of the most powerful to support that belief because it truly was a case that if you gave placebo after ECT, about 50 % would relapse within six months.
That other 50 would go on and have great lives. And you could reduce it markedly 20 % with drugs. What we were able, where at Columbia we were seeing is far higher rates of relapse. That relapse was a huge issue with ECT. And we didn't know what the true underlying relapse rate was because everybody was being treated and so on. So, we were funded to do a multi-center study that we published in JAMA.
in 2001, which is the only modern study ever give people placebo following ECT because we didn't really know what are we trying to improve. In that study, we randomized patients who were remitters to ECT to either placebo nor tryptaline alone, which it was a tricyclic antidepressant, very powerful effective antidepressant, or more tryptaline plus lithium.
Chapter 3.15: Increasingly Refractory Patients and Effect Sizes
And the relapse rates in six months were 84 % with placebo. If you're on after ECT, it's not the 50 % from the England, it's 84%. And what is important to realize in Declan McLaughlin, the chair at Trinity College Dublin, had a wonderful paper in neuropsychopharmacology showing that from the 60s till around 2015, the relapse rates with ECT have increased across time, that there's a secular change. And my interpretation, and there's, think, a lot of support is because the patients are more and more treatment resistant. That's the power that we are selecting for more and more for the people who are going to do badly with the treatment or with relapse. And in fact, the amount of ECT we give in the world has also almost doubled or tripled for many patients in terms of how many treatments they get over time. So when Kalinowski, who introduced ECT to America at the Psychiatric Institute, said, how many patients does a depressed patient need? He was the one who first identified its remarkable efficacy in depression. was four. Four to six was the median number of treatments that depressed. Then it moved up when the drugs came out to six to eight, then eight to 10. And then now it's in the 10 to 12 range in the average. And some patients…We took a group of patients, we just had a paper and brain simulation on this, but we took a group patients who failed ECT and gave them a second course. And we had a 70 % remission rate. And even when patients got the same type of ECT twice, that basically you need a lot more ECT today than you used to use in order to get people well. And I think that that's a secular change due to the nature of the patient population, the resistance of these people.
Marom Bikson
And is that because patients have, let's say other options like TMS, so almost by nature, once they're making it all the way to ECT have failed ....there's all these new treatments. So now they have to fail all these to make it to ECT.
Harold Sackeim
Exactly. They had many, many, many more options. when we first did VNS, when we were one of the four groups to do the first implants with VNS for depression, my patients at Columbia averaged, I think it was 20 failed drug trials and all of them failed ECT with me before getting VNS. So, you can imagine these are really, really tough people. you know, that…
So, it matters who is being looked, when you're evaluating a treatment, you need to know, you know, essentially what have they received before and how much they failed. So from the relapse point of view, ⁓ what we found essentially is that when you add lithium to an antidepressant and another subsequent multicenter trial, we added lithium to Vimlifaxine and had the same effect that pharmacologically that's our best strategy for relapse prevention. It's not great. mean, 40 % is a lot better than 84%. But you can also use ECT itself. And Charlie Kellner's work with the core group has demonstrated that when you continuation ECT, a rescue ECT, you can do even better.
Marom Bikson
Very very difficult.
Harold Sackeim
And that's probably that's, I think, true. And, and in the future, what I'm hoping we're going to see is whether you could use TMS as, as, yes, that there are great opportunities here to promote the stability of the treatment. And, but the larger point is, you know, when we did the work with durability with VNS,
Chapter 3.16: Rescue ECT, Maintenance Stimulation, and Durability
Harold Sackeim
Durability now has become, for some of our interventions, the reason for existence, the reason we're interested in treatment like VNS is not because of not, know, maybe let's say 40 to 60 % will show some degree of benefit after a year to two years of getting this thing. But you're not gonna go through surgery to get a modest level of improvement unless it lasts.
And what the VNS data showed was that for almost everyone at last, it has remarkable durability and benefit, even though they're a highly resistant group. It's sort of like very similar to what we see, think, with Parkinson's disease movement disorder phenomena and DBS, that you get great persistence of the effect in a population that has done very badly with drugs.
Marom Bikson
So, going back to this question of why you relapse, right? That way you're explaining it, and this, again, is not specific to ECT. This would be any antidepressant. Is antidepressant treatment. That depression is emerging in episodes. And what essentially you're doing is you're truncating that episode so it's over. However, you have not necessarily treated the underlying disease. And therefore, at some time point, another episode will inevitably reemerge?
Harold Sackeim
Yeah, many of these illnesses are recurrent. But when I started to give you the sort of the full picture, when I started, it was very common to see people who had a fulmination early in life. had an acute, severe episode, maybe psychotically depressed. You get them well. And then I saw them 30 years later after they had a great life. And now they had another recurrent episode. And nowadays,
So that is where you saw depression is like the flu. It was episodic and unless you fully treated it, but nowadays the much more common clinical presentation are people who've been sick for 10 or 15 years and they're sort of half well and they're being frozen in place in many ways with the pharmacologic agents.
And I don't know whether their extent. So, our models have gone from being a episodic flu-like phenomena to more like a diabetes phenomenon. So that once you're on a drug, you're on the drug for the rest of your life. And the illness is now being conceived as a chronic illness. So much more like congestive heart failure than anything truly episodic. We don't talk a lot in the field as to why our underlying models have changed. And I think it's..again, I think it's adrogenic that oftentimes because the pharmacology is there. It's not that for many of these, in my own work, I did multi-center work where I showed that if you gave people placebo during ECT, an antidepressant placebo, or a venlafaxine or nortriptyline, you started the drug, the antidepressants early, you get a 15 % improvement in your remission rate with the drugs than you do without them that ECT benefits from having the antidepressant on board, right? But it's also possible that those drugs are the same things that are preventing the resolution of these episodes.
Marom Bikson
Yeah. What are the things you're saying so far about, you know, the episode, maybe ECT truncating the episode. I understand it's more complicated than that. So now I want to ask you this question really beginning at mechanism, right? What is the therapeutic mechanism? Because generally speaking, right, we know that the all-neural brain stimulation works not because the brain lacks electricity, right? And we.... and there's like electricity deficit and we're putting it in. I would also say the brain doesn't lack a seizure, right?
Chapter 3.17: Wired to be Depressed 3 - The Circuity of Depression
Harold Sackeim
I joke at lectures and say nobody ever got depressed because they lacked a seizure.
Marom Bikson
So, I think the concept may be more that either the seizure or the electricity itself is activating a process that is built into the brain itself. There is a system in the brain that is a very complicated system because depression is extremely complicated and mood is extremely complicated. And either the electricity or the seizure is activating that system in a way that the patient cannot activate themselves. And then that activation
is leading to it. And, and for me, you mentioned them in the last part where we discussed it, how ECT could even treat status epilepticus, which is so weird, right? Why would a seizure treat a seizure? But that would also make sense if the seizure that you are producing activates a corrective process in the, in the brain and anti-seizure process, right? That then shuts down.... So, I wonder if you can kind of build on these, on these
Harold Sackeim
Yeah, my modeling has always been that we are wired to be depressed. Very different than my colleague, Don Klein, who I mentioned last time, the father of pharmacology, who saw depression more as a deficit state, right? So is the disturbance in appetite, the sleep disturbance, is it more like an aphasia or a release phenomenon? So when you're really thinking circuitry….the depression where we've produced, you know, like some injury to the brain so that it can't function the way it's supposed to, or is the depressed state itself an organized phenomena involving, you know, many, many at different levels, cognition, mood and appetite and vegetative function and so on. And I was always the one that believed that, both depression and mania are things that you can trigger.
Marom Bikson
And by the way I would tie that back to the very first part of this interview where you were talking about your early work about trying to understand how the brain is juggling, you know, happiness and depression. Go ahead. Yes.
Harold Sackeim
Right, right.
I'd always thought that yes, and in a very Bolbian sense, you know, the analyst from England that essentially the attachment there is, I think that one of the purposes of our being wired to be depressed is it allows us to love. That if we did not experience loss, and particularly in child rearing and the evolutionary importance of that,
How would we have these little things that hang around for 10 years that need all of this nurturing and so on, all the sacrifice if there wasn't the cost to pay for their loss? so whatever its purpose, I think we are wired and we now are getting into circuit where really with stimulation technology, we can see that we can produce manic states and we can produce depressed states.
And so, the excitement of ECT, because we haven't spoken about the spatial localization it offered, what it taught us about circuitry, was that it offered an opportunity to essentially figure out where in the brain do we have to intervene in order to get this ⁓ antidepressant type effect. My own theorizing, came largely out of epilepsy work, where we know in epilepsy you get a surround inhibition at a seizure focus. ⁓ And that makes a lot of sense because if you have a broken pipe in your kitchen, you don't want to shut off the water that's flowing in your bedroom. You need to shut it off at the pipe. So if you have a focus that's the engine for a seizure, where it propagates to may be less critical than where it derives from.
And that was exactly my view is what happens in the ECT because all of my work with absolute quantification of regional cerebral blood flow and glucose metabolism and so on, and the EEG work that we did all indicated that we're getting highly regionalized effects and in particular lateralized effects with the comparison of the unilateral and bilateral. But we tied, we were the first to tie antidepressant effects to the impact on prefrontal cortex. And, you know, we had paper in 94 in what's now JAMA psychiatry showing in, I think it was about 60 people that the more we drove blood flow down an hour after the treatment in prefrontal cortex, the more people were likely to get well.
Marom Bikson
And we'll probably revisit the prefrontal cortex story when we move on to TMS maybe in the next part. really want to, what you're saying is so interesting, so I want to sort of revisit it. So we are wired for depression in the general sense. We could say, you know, our brain has a circuitry that, a circuitry that is built to push us in that direction. Right. Would you say we have other circuitry that is designed to counteract that obviously and produce happiness and than…Is the in depression that what is the depression circuitry too strong? That happiness circuitry is too weak or is that not the right model?
Harold Sackeim
Yeah, the - it could have been a whole bunch of ways. Essentially, you're asking is if you have a circuit that's activated that results or is expressed as depression, is there something else that's inhibiting it or is it just inhibition within that circuit that's modulating it? and I think there's some reason to believe that there's reciprocal inhibition in the adult between
the euphoric states and the dysphoric states. That to me is less critical that they have separate representation. And as we discussed last time, think there is very little that's phenomenologically different about somebody who is essentially kicked in their soul by their spouse saying they're going to leave them or their employer saying that they're doing a terrible job and so on - and what you see in a depressed patient. I think that normal sadness has tremendous overlap. It's the same circuitry and what's abnormal in the depressed state is that it doesn't stop. And so that was always in my mind. Are we inhibiting this type of circuitry? It was my immediate mechanical view as to how something like ECT was working.
Marom Bikson
Overactive depression circuitry letting the system reset in a better balance.
Chapter 3.18: After the seizure
Harold Sackeim
And that the only reason we were producing seizures was to get the brain to stop them. That was also, I think, conceptually an important contribution because the focus of the field through the 40s and 80s was on the biochemistry that occurs during the seizure itself. And the state of the brain following the seizure was really not looked at consequentially.
And so, it does prolactin release during the seizure does it really and many, many, many things. And. And so the discovery that ECT was an anti-convulsant that it induced slow wave activity, Delta activity, EEG, which also co-varied with efficacy that it resulted in an inhibition of glucose metabolism. even though CBF goes up 300 % during the seizure, it's below baseline following it.
That all was consequential in my theorizing. Nowadays, and I would criticize our field strongly for this, the term neuroplasticity is the explanation we always give for why things work. It's because we're impacting on neuroplastic processes. That's sort of like insight and psychoanalysis. It's sort of like a meaningless.
Marom Bikson
That means something has changed, okay.
Harold Sackeim
… ECT is the strongest promoter of neurogenesis of any intervention we have. We did the work in primates, demonstrating that ECT promotes the development of new neurons, functional, active new neurons. But that process takes months to become functionally relevant.
And it's totally out of the time course of ECTs. You get a single one treatment and you're getting almost a 30 % improvement in depressive symptomatology on average. And certainly, the time course that you would take for having the full neurogenic effect. Now it could be other, certainly other aspects of neurogenesis, many, many other aspects like synaptic density and have a very different time scale.
But we have no idea at the concrete scientific level what aspect of brain change is tied to the efficacy of this treatment, particularly in the long run.
Chapter 3.19: Mechanisms of ECT , Electricity or Seizure?
Marom Bikson
So let me ask you two more questions to wrap up this part. And the second one, I think we'll set up for the TMS discussion to come next. And they're related maybe. That first question is, in general, across brain stimulation, this is true for ECT. Stimulation changes so much in the brain. You zap, this goes up, that goes up. Blood flow goes up. Therefore, there's going to be a response in this way and that way. And the risk of that is almost
It may be the case that almost anything you look for will change. You'll look for this, you'll look for that. GABA goes up, glutamate goes up. And so on the one hand, you're finding all these mechanisms. On other hand, you see my concern. It's hard to understand where the specificity is. Because again, you produce a seizure. I'm pretty sure everything changes. And then I can also just tie that to the second question, which is where I want to end up is, is it the seizure or is it the electricity?
Harold Sackeim
In terms of the first question, that was, that's the hopelessness that Seymour Keddy expressed very famously and I've written somewhat about it. Keddy was one of our great psychiatric scientists of all time. He was the one that invented the Keddy Schmidt method, which was the first method we had for measuring cerebral blood flow in the human being. it was a, it was a, it…
I won't go through it, but it was essentially you had anesthesiologists that were doing very invasive things to the carotid artery and the other in order to measure the whole brain. And by the way, the first group of humans that were studied with it were patients getting ECT, patients with schizophrenia getting ECT, where he demonstrated reductions of blood flow with the originally with the Kettyschmidt method.
Kedi then went on to do the first twin work in schizophrenia. He was also one of the ground breakers in discovering the catecholamines and their role in the brain. So Kedi was a brilliant, brilliant, expansive mind in our field.
Marom Bikson
By the way, it's so common in general in science when you have someone who's such an original thinker, ...the hits just keep on coming. You know what mean? just a trail... and they're not even necessarily in the same domain. They just seem to the way their mind works. But yeah, go ahead.
Harold Sackeim
Very generative, yeah. And Kedi said at a conference, the book came out, if it was in Puerto Rico, the conference, that ECT does thousands of things that are consistent to the brain. And that the idea of ever figuring out what's epiphenomena and what matters for efficacy is hopeless. He's essentially expressing the view you just did. And my retort to that was,
Well, we now have the possibility of using forms of ECT that don't work versus forms of ECT that do work. And that subtraction is fundamental to our figuring out mechanisms. And so that was one of the great contributions of the low dose right unilateral forms of treatment because
When we reported in JAMA Psychiatry what the physiologic differences were between people getting well and not well, was between ineffective treatment and ineffective treatment. I don't believe for instance,
Marom Bikson
And in effect, it was also producing seizures. So was not just that.
Harold Sackeim
Right, it wasn't just that, we used conservative criteria for what constitutes a seizure. But and I argue that if you give Haldol or Risperidone to people, the amount of D2 receptor blockade that they're going to get is going to be totally independent of whether the drug works or not. For some, it's going to be relevant to their pathophysiology. For others, it's not. But when you give powerful treatments - the idea that they're going to work differently in different people is, and that's 95 % of our mechanistic studies are correlations between outcomes and some biological measures and people who receive very powerful treatments. so that I think you have to weaken the treatment in strategic ways or change the treatment strategic ways to ask the mechanisms of action of that treatment. Otherwise, we're just doing correlational studies with what gets better and people who get better or not. It could be many other things. I think we've had that opportunity.
Part of the reason there's very little ECT research in the world, far less than there ever should be. Unfortunately, there's a dearth of it. And on top of that, to be able to do experimental forms of ECT and to use it in a way that you know is not going to be as effective and so on. You know, so you can see what the limitations are pursuing that line of thought.
The other big limitation is there's no good animal model for dosing in particular. You cannot find the seizure threshold of a rodent because I spent two years trying to develop models where we could do this in rats. And they had many site seizures, which are dosage related. So that, for instance, when you do kindling, ⁓ there's a scaling of the seizure expression with the kindle stimulation. And so, you can say, at what level did I get a seizure or not? And if you get one event, you can try again.
So, nobody has effectively, the human is much more simple. So is the monkey, by the way. Monkeys are incredibly reliable in terms of what their seizure thresholds are. ⁓ And like the human, it's almost all or none because you get this propagation to the motor cortex and you see it. In other animals, it's more complicated.
Marom Bikson
Sorry, got it. You said something here. I need to take this a little bit of a tangent here. You talked about kindling as a concept that you could use electrical stimulation sometimes above seizure threshold in order to make the animal more prone to now epilepsy. So then you say, ⁓ therefore can intense electrical stimulation, but that's not what we see in people. In fact, if anything, we see the opposite, right? That it's anti-convulsant. there is a bit of a, there's a, it's, we're doing, it's kindling in animals, but
Harold Sackeim
If you give ECT to an animal before you kindle them, you block the ability to develop kindling. So, it's like many processes where once you exceed a threshold, it becomes a very different phenomenon than below that threshold.
Marom Bikson
You need to get to the seizure for it to have the anti… if you just…
Harold Sackeim
Exactly. At least in my theorizing, the seizure triggers the process to shut itself off. And you don't get that necessarily with just intense electrical stimulation.
Marom Bikson
Okay. And then the second part of the two part question, and this, we're not going to wrap this up in this part, but I just want to get the juices _ flowing. And then we'll, we'll, get all into TMS and other things at the next part of this interview. Is it the electricity or is it the seizure? Meaning if you could, you certainly classically, right? You can produce seizures without electricity and get some of the benefits. So that might argue it's the seizure. Alternatively, you know, is it a crazy idea to suggest that you could just apply electricity - and not produce a seizure and still have antidepressant effects?
Harold Sackeim
I think both are true. don't think, and this again, this, I think very narrow thinking that has to be one or the other was what produced the warfare between the ECT world and the TMS world and the rest of brain simulation. There's a very cold reaction to the development of TMS by much of the ECT world - not because it was necessarily competitive, but it was ideologically heretic. And in fact, at one meeting of the ECT society, essentially ISCN at the time, I think it was ACT then, Mark George was left out of a meeting because he was suggesting that nonconvulsive stimulation of the brain could have antidepressant effects. Here, in a more nuanced way of thinking about it,
I developed a technique called Focal-electrical-administered seizure therapy, FEAST, which aims to essentially spatially target the electrical for ECT. that we also, I was funded by NARSAD for established investigator award for FEAT, which is without the seizure : Focal-electrically administered therapy.
And it's quite clear that intensive electrical stimulation does many, many things that could be therapeutic in the brain. And as one example, ECT is anti-Parkinson. There have been sham control trials where it's quite clear that you get a reduction in the motor disturbance of Parkinson's disease when you deliver ECT. And at one point in time before the more widespread use of clozapine, maybe 20, 30 people a year were referred to us at Columbia for ECT, the part because they were psychotic from their drugs, the L-Dopa and the dopamine agonists and you couldn't give them anti-psychotics. It was thought because of the dopamine blockade, you were Parkinsonite, and worse. So ECT was the solution because it was both anti-psychotic and anti-Parkinson. Well, it turns out that if you do microdialysis of the rat brain, you see that ECT results in a massive release of dopamine among other things in the basal ganglia. But what you do is if you block the seizure with a barbiturate and do the same simulation, you get the same amount of dopamine release. If you turn the dial up, you get more dopamine release.
And I use that clinically when treating these patients with Parkinson's disease, because it was, you know, turn the dial up and the patient will come back and say, whatever you did last time, it lasts twice as long in terms of my motor symptoms as you know. So there, I think the electricity has a great deal to do with the clinical impact. That doesn't mean that the seizure also doesn't have its own levels of contribution that.
And so, I've always been sort of…of the mind that it doesn't have to be one or the other.
Marom Bikson
Okay, so I like that. Let's leave it on this FEAT This strange idea of focal, Focally administered electrical therapy. That's gonna set it up for the next part where maybe we'll talk about the birth of TMS as you're saying, almost like in the shadow in some extent or in whatever it is of ECT.
Harold Sackeim
Alright.
“Wired for Depression” with Harold Sackeim, Part 4: Development of TMS
Chapter 4.1 : Wired for Depression 4 and What it Means for Treating Depression
Marom Bikson
Okay, Harold, as we continue sort of our discussion about your decades of experience developing neuromodulation with a special focus on depression, this is a part I'm very excited about to really talk about the birth of TMS and how you describe it in the shadows and the wings of ECT. And so, think it bridges very nicely with what we said before.
But before that, I actually wanted to ask you about something. It's something you mentioned in the previous parts of this interview. And you've told this to me before. And I'm sorry to say that it took me a couple of you know, sometimes you get an idea and it has to kind of enter and it doesn't the impact of it doesn't hit until you're like, you know, you're about to go to bed and then you're like.... you said something very profound that I want to kind of start with going back to it, which is that.
You talked about depression as something that we're wired for, not as a pathology, but something that is actually an evolutionary advantage. so can you restate that idea? then what does that mean as far as treating depression? then we'll go and get into TMS.
Harold Sackeim
Yeah, I think fundamentally from a neuroscience point of view, either you're a release phenomenon when it's something this complex and integrated, or you're a deficit phenomenon, or to some extent you could be both. But that's from a neuroscience point of view, and that's fundamental for circuit thinking. And there was good reason for people to think that the depressed state reflected was like aphasia the interference with ongoing function because it resulted in deficits and movement and appetite and sexual activity and so on. So, what you were doing in terms of restoring function and using that type of thinking would be very different because then it would be a rehabilitation type of treatment.
Marom Bikson
In that model it's almost like an injury that you need to heal.
Harold Sackeim
Right, exactly. And the other model, and this I think runs through a lot of our thinking about pathology, and is less seen often in neurological circumstances, is a release phenomenon like a seizure and automatic speech.
And so, the issue was, is this a affective organization, a normal state that's gone wild in some way, uncontrolled or uninhibited, or was this really a disturbance in the expression of a, you know, this state? And obviously it could be both, but that was essentially my thinking was, and the phenomenology, I think, was- was there and also what seemed extraordinary about mood was that the triggers could be abstract, propositional, that ideas could trigger mood, That what we thought mattered in terms of our feelings. And that doesn't have to happen biologically.
And so that made it a two-way street - that our moods distort our thinking and our thinking can distort our moods. And that's where psychological treatments essentially have a space. And the biological level is always being expressed in terms of some circuitry, at least that's how I've been thinking about these phenomena. And in ECT, what was profound there was that you could set a dial on a machine, you know, change the amount of charge you are going to deliver and have 10 % of people get well or 70%. That was a proof in your face, so to speak, you know, in life that the underlying circuitry that was being altered was fundamental to the experience. you know, it could be modulated that way.
And of course, that's always been one of the...
Marom Bikson
We say proof, you mean that there was a biological substrate. Yeah. Obviously the stimulation wasn't talking therapy. The stimulation wasn't asking them about their mother. The stimulation was going to their brain.
Harold Sackeim
In a positive way, you could alter people's mood and that the circuits go both ways. Thoughts can change mood and the biology of mood will change the thoughts too. And in this complex world we live in. But that simplicity because you could do a real experiment and figure out circuitry was very powerful for me in terms of the understanding that this has much wider implications for the field of neuromodulation.
Marom Bikson
Even going back to what you initially said, even from a circuitry perspective, I guess you could imagine it as we have a depression circuitry. And it has now been, like you call released in a perhaps pathological way, versus we have a circuit and now that circuit is broken. And that's why we're…we shouldn't be sad, it's broken. And that's why we're sad, right? Those were that you still have that, that aspect of it that you're talking about.
Harold Sackeim
Right. And the fact that you could inhibit circuits, that in fact, that was our language and our thinking at the time. It was very simplistic in many ways. And it was the work I was doing with some of the first circuit based thinking in terms of how it underlies emotion and how there's breaks and so on. But that all predated the connectivity. perspective. It was essentially the same type of sort of mechanistic approach. And I think it has a real reality. mean, it's actually achieved some proof.
Chapter 4.2 Birth of Biological Psychiatry and ECT
Marom Bikson
All right. But now, that's very interesting. Okay, so let's talk about TMS. Let's talk about electrical stimulation that does not produce seizures, yet can treat a variety of disorders, including mood disorders. So, you know, from your perspective, how did TMS emerge? What were the big milestones?
Harold Sackeim
Well, yes, a little bit about that. In many ways, the biological understanding of the major mental illnesses was the original academic approach. I mean that's where we have the Creplins and the Alzheimer's and that real tradition. It was a neuropathological tradition. It was a tradition that the fundamental perspective of that field was that the major mental illnesses, both mood disorder and psychosis were either degenerative or they were congenital. That there was nothing you could do to alter the path of these illnesses. That you had no, no, there was therapeutic nihilism.
Marom Bikson
…for mood, including for mood.
Harold Sackeim
And the average length of stay was two years for hospitalizations from mood disorders and so on. were profoundly impairing, you can imagine, bipolar disorder in particular. And it was ECT that changed the whole world in terms of the perspective that you couldn't do anything because with the advent of ECT, tens of thousands of people who were institutionalized were discharged went back to the community. was like night and day for many of these... This is in the 40s, the late 30s and 40s. And it ushered in the pharmacologic revolution. Right. And it was ECT _ in fact, that the antipsychotics and the antidepressants were tested against to see how did they stack up in terms of, you know, right, you know, and
Marom Bikson
And when is this?
Interesting.
Harold Sackeim
And then interest in ECT itself proceeded academically, very little with the introduction of the drugs. was sort of like, but it was the drugs that because of their incredible widespread use. mean, when you think of it, that the percentage of the American public that is on a psychotropic is quite extraordinary.
And the fact that it was the introduction of the drugs that ushered in a modern science of clinical trials. And this is across medicine. That was one of the impacts of the psychopharmacology was that the need to develop reliable diagnostic and outcome assessments and the importance of randomization and many of the things we take for granted across all of medicine really was spurred on by the advent of the psychotropic medications and NIMH establishing units, assessment units basically, ⁓ to promulgate modern scientific methods.
So, I think that also was important in spurring this biologic revolution as well as, know, from the scientific point of view, the discovery of the catecholamines and the underlying biology and so on that created a whole different framework. And American psychiatry was, you know, sort of had this battle with a psychoanalytic tradition, you know, starting around the 20s and 30s, because that became a very powerful voice within psychiatry of many of the major departments essentially being totally psychoanalytic in its expression. And that was to many extent true at Columbia and where I was. And just before I came, the new chairman of the department who was Ed ⁓ Sacher, who had discovered the first biological abnormality in depression, phase advancement of the release of cortisol. He did the first 24 hour urine samples and so on. He and Don Klein essentially fired all of the the analysts and transformed that institution. So those types of battles went on in many, many places and the field went over in that direction.
Chapter 4.3: The Origin of TMS - A Technology Problem
Marom Bikson
Okay. All right. So now, that's very interesting. Okay, so let's talk about TMS. Let's talk about electrical stimulation that does not produce seizures, yet can treat a variety of disorders, including mood disorders. So, you know, from your perspective, how did TMS emerge? What were the big milestones?
Harold Sackeim
The very early history of TMS is fascinating, and I think the highlights are pretty clear. The idea of trying to induce current in brain without pain had been around for quite a while, and some of the very important basic work being done by Merton and other people to map out the motor system basically, how to use painful electrical stimuli.
Marom Bikson
So again, this would be electrodes placed on the head, connecting it to a stimulator that can produce an amp. But unlike ECT, you keep the frequency low so you're not producing seizures. But you can do things like produce these involuntary motors. But I've had it. I mean, I think you've had it too. It's really, it's torture level painful, right? It's rough.
Harold Sackeim
Right. there's no doubt that ECT simulators can be used for torture, that they're capable of because of the magnitude of the stimulus that they would provide. So that was an incredible…
Marom Bikson
to clarify if people are asleep so that there's no pain….
Harold Sackeim
During ECT itself, there's no pain at all. Right. And that was always a huge impediment to the study of stimulation is that you can't get enough of it in the brain to do anything with that. being painful in the awake individual and back in going into the late 1890s, there were attempts to generate sufficiently powerful magnetic fields by Maxwell's equations to induce an electric field that would do the same thing. And they were successful, you know, decades and decades ago in producing phosphenes, ⁓ which is likely to the fact that the retina has extremely low thresholds.
Marom Bikson
The phosphenes would be flashes of light. And again, it's in a sense, it's not getting, it's still not transcranial because it's still peripheral right? You're just getting, just activating.
Harold Sackeim
It did something, but, and the weakest possible thing that you could do. And that was the, it was the electrical engineering problem, which fundamentally in many of the things that we do, that is what the thing that we come up against is a field. And that's where the progress has really been moving the field is at the technology end, is giving us more and more powers to do things that we never thought we could do.
And Tony Barker's instantiation of that at Sheffield was a really brilliant illustration of the persistence and the innovation of a single individual who was not in an academic setting, who had profound respect for an interest in the history of science and thought it was possible. And so, he did it.
And this is exactly, my understanding is that Tony was a physicist in a hospital setting and he had an academic appointment. But this was not his primary labor. And for what it's worth, I recently was asked to write a letter of recommendation for Tony to be knighted…was very happy to do so.
Marom Bikson
Electrical stimulation transcranially going through the skull too painful. Barker builds really the first modern example of a transcranial magnetic stimulation device, a TMS machine. This is just one pulse at a time, right? But it does the trick in that where, now for example, you can produce involuntary motor twitches by stimulating the motor cortex, but without the pain that is associated with the electrical approach, right? And that's this is in the UK. that's low frequency, but you're proving that you can get through the skull now.
Harold Sackeim
Yeah, and what was brilliant and to me of great interest about that approach was that you had a greater control over the electrical field than you would with transcranial electrical stimulation like an ECT. That going back to the ECT work, my work had strongly indicated that the modulation we were doing of prefrontal tissue was fundamental to the efficacy. that ECT didn't work when you were triggering seizures from the motor strip, for instance. That's how low dose rate unilateral, my theory has always been, is the reason it was ineffective is that it was the wrong site and seizure triggering. And that the motor cortex has the lowest threshold in the cortex for seizure activity. And so, by lowering the dose there, that's where we're getting it from.
And there had been work in the 40s in ECT doing bi- occipital which resulted in visual defects, as you would expect, and so on. And so that the TMS offered a tool to do the same thing, but to do it with greater precision of dose because of the cleanness of it not being essentially distorted by the scalp and the skull. _
So that was the beauty of invention to me.
Chapter 4.4: Repetitive TMS, from finger twitches to excitability modulation
Marom Bikson
So just so now the next step is instead of applying a single pulse, apply trains the pulses to tick, tick, tick, tick, And depending on the rate, the frequency of those pulses, which could be faster or slower, there was evidence that would suggest that you could either increase the function of the brain, whatever that means, increase excitability or depress brain function. So now you had an intervention that could sort of tune brain excitability.
Harold Sackeim
Right, that, and you know, it's interesting that that was a metaphor and there was exciting data that was somewhat in support, but you don't see a continuance of that approach. And instead, the metaphor has now shifted to connectivity, which, and just as an aside to me, I've been a critic of the connectivity stuff, even though much of the imaging work I did was among the first to look at the covariance patterns and ask which regions speak to which regions and so on. Largely because of the nature of the measures that are being taken with fMRI in the bold response is so limited in its dynamic range and there are many factors that will impact on dynamic range that looking at the correlation between regions when you only have a plus or minus two percentile dynamic range seems to me very, very artificial. So it's one thing to say that blood flow has decreased in a region and another thing to say, well, its correlation with another region seems to have changed, but- you know...
Chapter 4.5: TMS development - The leap from changes in excitably to treating depression
Marom Bikson
All right, so be as it may, we now have TMS machines. We have TMS machines that can have some rate. Tick, tick, tick, they can go to it. And we have the idea, right or wrong, that depending on how quickly you set that clicking rate, you might be able to even increase brain excitability, whatever that means. But that still is not....
Harold Sackeim
Right. Or decrease it. Right. That you had demonstrations of two things with that. You had experimental demonstrations that you could do things like speech arrest interfere with - get a twitch, but, but, you know, essentially do functional mapping with it. And so that made it look like it had, it was capable of modulating important activity of the brain in a way that had, you know, behavioral outcome. And, and you produce seizures that
That to me was a highly attractive adverse event to the extent that under some conditions, this technology was capable of taking normal individuals. - these are often subjects in psychology experiments who had a seizure so that it might be useful for the deliberate induction of seizures as well.
Marom Bikson
So, this was an adverse event, not a single pulse TMS, but a repetitive TMS. And even to today, this remains the main concern with TMS. And a lot of the safety protocols that exist around TMS are designed to sort of as robustly as possible prevent the induction of seizure. So, we can get back to doing it on purpose, right? But the general framework remains TMS, multiple pulses and very intentionally do what you can to actually avoid seizures, right? So, it's a sub seizure intervention.
Harold Sackeim
That’s the hope of it. Or it or it could be above or sub. It's an issue of electrical engineering. What are you giving.
Marom Bikson
Okay, but still how with all these pieces in place to decide that this could be used to treat depression requires a little, still there's a leap that has to be made there, right?
Harold Sackeim
I think the leap that Mark made, because he was really the first to not only study it, or he was among the first, I should say, because there was a host of labs that were doing this at the very beginning. Basically, the field of TMS went to fast left, slow right dorsolateral prefrontal cortex targets. And they were very fortunate because both work.
Marom Bikson
Hm.
Harold Sackeim
And both worked against sham. mean, you know, to this day, if you want to say, where can I reliably know what the target would be for depression? And they and they were crude in how they got there and some we could go through all of that. But it was an incredibly fortunate set of choices. And there were bunch of choices that were made in this.
Marom Bikson
So there the notion is frontal left, meaning fast higher frequency in order to try to enhance the boost. Right slow, meaning lower frequency, supposedly to depress. So, the field moves in this direction. But who thought that that would treat mood?
Harold Sackeim
Well, there had been, remember, as I said, when I started ECT work, ECT research, it was with the idea that there was a fundamental lateralization to the nature of emotion and largely that work and the work of Davidson at the time, this is, we're talking about major papers are in 78, at the time had raised the hypothesis that hyper excitability on the left was associated with euphoria and on the right with depression and that you could inhibit either of those to treat those states. And that was in fact the type of thinking that brought me into studying ECT originally. So it was..
Marom Bikson
And as we talked about in the first, I think the very first part of this interview, okay, this lateralization and…
Harold Sackeim
Well, that I don't believe it. I think that that was a gross oversimplification, but the interesting one. But…that infused the field of TMS from its beginning. And that stays to this day. And as I think I said in that previous interview, the fundamental study, the two by two, where you credibly go slow, fast left right and you randomize has not really been done. And I think it would disprove and be of clinical relevance mechanistically because then it would have to change totally that we think that fast or slow matters in these ways.
Marom Bikson
Okay, but still at the time, Mark George and a small group of people around him decide that there's an imbalance issue in depression and excitatory to the left or inhibitory to the right will correct this imbalance, right? That's sort of the state.
Harold Sackeim
And part of what helped with that is that there may have been, and it's amazing how we forget our history, maybe 20 or more PET spec studies quantifying blood flow or glucose metabolism, showing a negative correlation between how depressed you were and metabolic rate in the left prefrontal cortex, particularly dorsolateral tissue. It doesn't mean that it wasn't also on the right because a large number of studies also showed it there. my work showed that the bilateral decrease was associated with getting better, you know, the bilateral prefrontal and so on. So, there are a fair amount of things that are embarrassing to that perspective in choosing that and making it free. But it was a good choice. ⁓
Marom Bikson
Eventually it was... so OK, but so around the same time, and I think Mark George is one of the earlier people actually involved in fMRI and functional imaging. So, he's also seeing what he believes are these functional imaging-based evidence for some sort of lateral dysfunction.
Harold Sackeim
Right. And among the things that was convincing to me is that an hour after an ECT treatment, all the changes you've got in the reordered on cerebral blood flow and with absolute quantification were on the side that you simulated. We're on the right side. was no lateral ... Whereas you could produce lateralized effects galore in terms of the impact on memory and
and affect and so on, but it was clearly that you could do those things through functional changes on one side of the brain.
Marom Bikson
So, Mark George gets up one morning, someday, right? And he's thinking about all these things, all this history. He's aware of it that you brought up and this imaging work. And he says, let's start doing excitatory, repetitive, to the left side of the head. And that's good to treat depression. is that?
Harold Sackeim
Mark is a fellow with Bob Post at the NIMH and a research fellow. He had the very, very good fortune of being at Queen Square in London when Tony Barker lent John Rothwell one of his very first devices, very first to really show effects in the human.
And Mark was on an elevator at Queen's Square where some guy said, they just did the weirdest thing to me, make my fingers twitch by putting something in my head. So, he immediately went to Rothwell's lab and observed it and came back to the U.S. and worked with Bob Post. And Bob was the chief of biological psychiatry at that division at NIMH. I'm the chief of biological psychiatry at Columbia. So, we're sort of in sister places and that's.
It was in visiting Bob's lab that I met Mark the first time and I met Bob Greenberg, who is also a fellow and who has done great things in neuromodulation. So that's where we met. Mark was committed to the idea that you could modulate circuitry in this noninvasive way and that if you gave enough of…
Marom Bikson
…Okay, that's the first time you met.
…without seizures.
Harold Sackeim
And without seizures, yeah. And you can imagine that did not endear him to a good part of the psychiatric community that thought that ECT was not just the way, but the only way. And so that generated a lot of disbelief and...
Marom Bikson
So, the reception was not warm.
Chapter 4.6: Early reception of TMS for depression by ECT community and Moderate Efficacy
Harold Sackeim
From some of the, yeah, from some of the community. And, you know, that's sort of, I think, very common that we get we sort of become identified with our fields of study or our therapeutics that we specialize in, and that we unionize - it's them and nobody else. And that is very common, I think, in our field. So.
Marom Bikson
But the pushback from the ECT field, let's say from the ECT establishment against TMS was partially related to the fact that the seizures were considered essential and therefore a sub-seizure therapy could not work. Was that the crux of it or they just didn't like something new?
Harold Sackeim
The idea that some type of stimulation of the brain that did not result in seizures would have therapeutic properties was either rejected as impossible, it was heretical, or it was...very unlikely to succeed. Let me put it that way, a lot of skepticism associated with it. And part of it, I think was, and you see this to some extent to this day, I sort of had a middle of the road position there to the extent that in relation to ECT, there's no doubt that when TMS came out, it was therapeutically pipsqueak.
It did not have anywhere near the therapeutic. And I wrote editorial and biological psychiatry, you know, saying essentially that it was not yet at prime time.
Chapter 4.7: TMS matures, efficacy increases
Marom Bikson
TMS was not ready for prime-time?
Harold Sackeim
Yeah. And, but that wasn't the issue necessarily because it was, you know, was finding its footing in terms of who was it useful to treat and how do you use it to treat and so on. And when we did the first meta-analysis of the therapeutic properties of TMS, we published that with Bert and Lisanby. Essentially, the conclusion was this is a really interesting intervention.
It's clearly better than sham and has antidepressant properties, but they're incredibly weak. That was sort of the, and I maintained that position after TMS was approved by the FDA, even though I was a senior author on the study that led to its approval. And Mark and I were the co-PI's on the OPTMS study, which was the independent - industry independent -study. And in terms of depression, both of those studies essentially found the remarkable thing that they found is that only 5 % of patients were remitted with sham intervention, even though you were coming in daily for treatment, seeing somebody daily, they had the lowest placebo, you know, compared to any drug trial, it was lower, yet the contact was so much higher.
And the remission rate was around 15 % at the primary endpoints. And 15 % ECT is around 70%, at least 50 to 70 % remission rate in a much more resistant ⁓ population. so honestly, when looking at those outcomes, my impression was…it may be of help to some people, certainly because it's not invasive. We've done some of the early studies on its cognitive effects and found that there were no adverse cognitive effects. but I wasn't jumping for joy. And what has happened is that TMS, and this is a really important, I think, lesson has matured to a point where
It is as good or better, I would say, than any medication that it's it's and it's only getting better. It's a moving target. But right now, and with data on, let's say, 50,000 people who've gotten TMS, I chair the registry data for Neuronetics, and they have data now about 175,000 people who've been treated with TMS the outcomes are, you know, more than double what they were in those initial studies. And so, we're talking about 60 to 70 % responding and 35, 40 % remitting. So, it has become, you know, really a very useful treatment.
Chapter 4.8: TMS After Drug Failure - Or Frontline
And the last thing in this is…In epilepsy, if you fail two anticonvulsants, you're a candidate for surgery. Because we know enough about the sequential treatment that the likelihood of benefiting substantially from the next anticonvulsant is there. We're getting to that point in depression where the likelihood of sustained benefit after so many failed trials is very- Drug trials is very, very low. So, this, and of course the labeling for TMS and VNS and now as ketamine were the first times the FDA ever labeled anything as a level of its treatment resistance. All of these treatments have essentially introduced treatment resistance as a subgroup and how do you assess that and so on.
Marom Bikson
…its part of the label that you have.
Harold Sackeim
And we don't know where TMS stands there. I'm hoping to move it more and more as a treatment of first choice because of the lack of side effects. I think it's an accident of history that it was tested in people who have failed medications to begin with. So, I think its therapeutic potential is even greater. And then the last point is TMS is remarkably lawful, in a way that surprises me, that the efficacy as a function of how many treatments you've had and where in the treatment sequence you have, we're really now defining essentially the functions that are associated with clinical benefit with this intervention. And it is remarkably stereotypic. And so that gives us all sorts of, to give you an example, in a sample of over 7,000 patients, we showed now replicated in multiple other samples that when you start TMS, you're improving at a rate of about 3 % per treatment. Then after five to 10 treatments, it drops to less than 1 % per treatment and stays that way to the end. This is with standard TMS where you're giving it one session a day. And so, the big question is, why is there that big drop? Can you accelerate that? Can you rescue for that drop and all sorts of things that will define the future of this treatment. how it has matured in terms of being, you know, there are now more TMS providers in America than there are ECT providers.
Chapter 4.9: TMS and ECT for Depression - Constant Innovation
Marom Bikson
One thing you've told me I think is very important, I want to underline it is the fact that the TMS as it was born, that there was constant innovation and it remains that its constant innovation to make the treatment more effective. And so, the reason we have the TMS we have today is only because there was this aggressive and ongoing push to do it. I would sort of on one hand, would contrast that with areas like, for example, TDCS, where I think, unfortunately,
That push hasn't happened. And so, we're still working with the original, right? And that is certainly a missed opportunity as far as optimization. But also the contrast with ECT, we're talking about, think, the two parts ago of this interview, there was a bit different. You had the potency to begin with. Right. It was the side effects that you were optimizing to.
Harold Sackeim
Right, but ECT is a great example of where we did the reverse, that it started at high intensity, a stimulus that had to be constructed in order to make it safer. What happened with TMS was because of the seizure concern, it started off with the weakest approach to treatment and has slowly until then somebody like Nolan Williams has accelerated the field to use a pun because he said, I'm going to give it everything I got and see if I get a faster and more successful treatment. So, when he went from one session a day to 10 from 600 ITBS pulses to 1800 in those sessions on and on, neuro navigation, so piling it all on into one, that wasn't done or couldn't be done. And that's the same thing with TDCS. And I would think is that…that hasn't pushed the limits to know what the limits are.
Chapter 4.10: Early Reception to TMS
Marom Bikson
You have the therapeutic window right between no potency and the side effects in your titrating... And it's not ... the dose base as we're saying it's not just intensity, it's intensity, it's session duration, session number, personalized, all these other things. You're titrating that space to give you that therapeutic window. Okay, now. We moved ahead. Now we know TMS is a very effective intervention. You've just documented the reasons. That wasn't clear when Mark George and company were first proposing this idea. So, I just want to go back to that time point. If you could recount to me, you told me the story before about some of the pushback that Mark George got at NIH and describe that pushback. Also, what had to be done to overcome that pushback?
Harold Sackeim
Yeah, you know, sometimes when you are taking what seems to be an original path, others seem to have really pretty strong ridicule. And when Mark was a fellow at the NIMH, I was there a decent amount of the time because I had the job of being executive secretary for a blue ribbon committee that was charged with essentially reorganizing, recommending reorganization of the NIMH. NIMH was the first of the intramural institutes to undergo this type of outside review where everything about it, its mission, the quality of its research, its educational components, clinical components, budget, certainly were all under review by an outside... And we actually wrote a book on recommendations on what to do with this. And so, I was in Washington when the brand new director of NIMH, I won't mention who, one can look it up historically, but this new director, who I had a fair amount of contact with given the role that I had there, had an all hands meeting - his first meeting with the intramural program and it was everybody at the intramural program had to be there and it was in the auditorium there and I happened to be sitting next to this fellow by the name of George.
Marom Bikson
A young fellow at that time, right?
Harold Sackeim
A new director of the Institute announces pridefully that NIMH will no longer support idiotic research, research that is simply a waste of the taxpayers' money. And as an example of this research we've been doing, research he says that attempts to link mood to areas of the brain using imaging, and that's phrenology, total phrenology.
And an example of a total waste of taxpayer's time and the perpetrator of this fraud was a young fellow by the name of Dr. George. you know, so he marked with….
Marom Bikson
He calls him out for the imaging, not for the TMS.
Harold Sackeim
Not for the TMS, it was the imaging, believe that was the, and I haven't, you know, I knew Mark, we had discussed TMS at this point and so on. So, I turned to Mark and said, you know, this is not a bad thing, you know, everybody now is gonna read your work, they wanna know what this is about, and so your name is out there. But it…You can imagine it was it would be traumatic for anyone to hear that, particularly from the head of your institute. And as I understand it, Mark was not allowed to talk about much of his work. He was instructed not to speak to the press. I he was was muffled. and presumably that's One of the reasons why he ended up going to MUSC going back is because of the reception he received in Washington.
Marom Bikson
The first TMS treatments are being done at NIH, correct? And what was the reception?
Harold Sackeim
Bob Moser was extremely receptive of Mark's work, enthusiastic about it. And Mark had the advantage that Mark Hallett, a very esteemed neurologist, and in fact, Mark became our first deputy editor for neurology at Brain Simulation. was at the Neurologic Institute doing work with TMS.
Marom Bikson
But so, as these, I guess, is the first papers or maybe in the first abstracts are coming out of TMS for depression, do you remember what's the reception among the broader psychiatric community? Was it excitement? Was it, I don't know, attack, criticism?
Harold Sackeim
I don't remember colleagues coming up to me and saying, hey, did you hear about this thing called TMS that there wasn't a lot of jumping up and down? I think one has to remember that the initial work that was done was often small sample work. The early days, for instance, there are a slew of studies of 10, 15 people per group given active or sham for five treatments or 10 treatments, claiming some degree of difference in therapeutic effect, often of small magnitude. So, there was a lot of skepticism. Is this real? it not real? A lack of real discussion in the community as to why this might be, you know what the mechanistic level, how could you explain it?
And the ECT part of the community to this day sort of has not integrated it that well. So, the journal of the ECT and other neuromodulation treatments has always had sort of an ambivalent attitude about the other neuromodulation treatments, how they get represented in that journal and so on and at the meeting.
Marom Bikson
So, you're saying maybe the adopters of TMS are not necessarily the ECT practitioners that ...
Harold Sackeim
Now, I think there's a lot of overlap. When I go to the meetings, there's... and I think the whole idea of interventional psychiatry is being an umbrella that subsumes these different interventions is a wonderful reality that, you know, and at least in terms of my original discussions with Mark and some of the stuff we wrote about, my thought was, if I had believed that the reason ECT was so effective is because it's enhancing inhibition and in prefrontal structures and is doing that through an anti-convulsant mechanism, the triggering of the seizure to stop the seizure, believe results in release of GABA and all of that stuff. There's no reason why you couldn't do a similar thing more gently where a pulse results in an increase in inhibitory effect that you build up over time that I didn't ever think that there was only one way to get to that same type of end. And I think that that is sort of what was key. And the other thing that, again, my view is that no matter what intervention you're talking about, I still think we have to be remarkably humble and that we're only three or four steps beyond bloodletting…that even though we have these new effective interventions.
Marom Bikson
…These work to be clear, these work, these are working interventions, okay.
Harold Sackeim
Working interventions, but there are a lot of people who still are being left out in the cold. so that's why, for instance, when we started the journal, it was explicitly saying that we have to be agnostic to technologies. We can't get into these intersign warfares if my technology is better than your technology.
Chapter 4.11: How Scientists Invent Brain Stimulation, Trial Design, and The Role of Industry
Marom Bikson
All right, so in the next part of this interview, I do want to talk about the journal. I do want to talk about maybe some broader issues in other technologies we haven't touched on and the future direction. I want to finish maybe with one question that maybe sets up that as far as bigger issues. The transformation of TMS from an academic exercise to an FDA approved treatment involved the company, right? A company had to come in and invest and sort of not just invest some money but maybe be very intentional about the direction of the field and the type of trial you needed to run to satisfy the FDA.
What are your general thoughts about the role of industry, about the role of companies, specifically in brain stimulation?
Harold Sackeim
I think it's very difficult for a scientist to invent a molecule. But it is not difficult at all for a scientist to have profound impact on neuromodulation technologies. That's, I think, you don't see a lot of individual labs that produce a drug that ultimately benefits mankind. ⁓ And I think that if you're doing innovative work in the neuromodulation space, you can't not be involved with a company, that in one way or another, I spent 10 years designing with Magstim the first stimulator capable of deliberately producing seizures in the human MST. And with certainly MECTA and SigmaStim they...
Marom Bikson
ECT companies, yes....
Harold Sackeim
The ECT company, I had custom devices made by them and we did RCTs with them, many of them funded by NIH and that immediately fed into what was available to the clinical community. And when we did what we now call the O'Riordan study, the pivotal study with Neuronetics to approve TMS, that was where the medical director at TMS, Mark Demitrek, wanted to import much of the methodology we used in ECT clinical trials. It's non-trivial about how you, for a procedure, how you organize your trials and what you do in terms of maximizing reliability of measurements and how you evaluate durability of benefit. And it goes on and on and on.
And so, if you want to try and give that technology the best shot and the most rigorous shot, which is sort of, should be the aim of companies as well, particularly if you're going to FDA, then that's largely going to come out from the collaboration with academia who have this level of expertise.
Marom Bikson
So, collaboration is what you're saying. It's not academics, it's not the industry, it's the collaboration that pushes things forward.
Harold Sackeim
Absolutely. Yeah, that you try and make, mean, it's hard to be agnostic and also collaborate with particular industries, but that I think is the attitude you have to take that you're playing it down the road and you want, you know, calling balls and strikes as they are and that's also the goal of the company that you're collaborating with is that they want the outcomes and the most rigorous way possible. When we did the optimization of TMS with Neuronetics, I'm sorry, with Neuronetics and the James Long company, what we did, our contribution in that trial was to do the first active of sham to use electrical stimulation to produce the same pain under a TMS coil, but without the TMS coil discharging so that nobody would know whether they're having a discharging coil or not. And so, in that trial, we tried to markedly increase the rigor of sham control trials in TMS using the same auditory input to the treater as to the patient, blocking any of the external sounds.
With Vahey Amasian and Mark George in my basement lab, we figured out how you can send an electrical stimulus, the pain associated with electrical stimulus to a particular spot. And so, we added that to, and Neuronetics collaborated with us because we had the MECTA Corporation, the ECT company build a stimulator that went down to two milliamps, what TDCS gives, right?
Marom Bikson
Yes.
Harold Sackeim
Right...We were in the 0 to 10 milliamp range with a MECTA stimulator
Marom Bikson
…for the sensation part, for the sensation.
Harold Sackeim
But it was triggered by the discharge or not of going to the Neuronetics. So, the two companies have devices that spoke to each other. So, that was where having positive relations with both companies resulted, I think, in benefits for the field.
Marom Bikson
Okay, I think it's a perfect place to finish this part and we will resume with a bigger discussion of where the brain stimulation has gone and where it's going and the foundation of the journal.
Harold Sackeim
Okay, great.
“Wired for Depression” with Harold Sackeim, Part 5: Targeting the Future, VNS, DBS, drugs, BCI
Chapter 5.1: Bottom-Up Vagus Nerve Stimulation vs Top-Down ECT/TMS
Marom Bikson
Good to see you, Harold. So, I'm excited to sort of wrap up this last part of the series of conversations we've had about your foundational experience in ECT and the development of TMS and more broadly in brain stimulation and specifically brain stimulation from mood disorders. So, we started talking a lot about the early history of ECT and then your foundational contributions.
We went on to sort of discuss the development of ECT, again, keep this in the context of some compelling theories you have about what depression even is. And then we moved on to essentially electrical stimulation without seizures, right, is TMS in its origin. So, I wanted to maybe start today by asking you about some other neuromodulation techniques that have been looked at, including for mood.
And so maybe start to talk about with VNS, if you can say maybe what your perspective is on VNS, where it's come from, where it's going.
Harold Sackeim
Yeah, I first learned about VNS at an international epilepsy meeting in The Hague, where I was giving a keynote talk on the impact of seizures on cognition. essentially, I was speaking to an epilepsy audience about the effects of seizures in ECT on cognitive parameters.
And I attended just out of interest, a session on vagus nerve stimulation, which was just emerging as a clinical tool to treat, intractable epileptic conditions. So, patients who had, treatment resistant epilepsy. This is in the mid-nineties…and this was a radically different approach.
Marom Bikson
When is it? Is this in the 80s?
Harold Sackeim
because it was clear, number one, that they were getting seizure control, that they profound anti-convulsant effects, but was also clear was this was a technology where you're wrapping an electrode around the vagus nerve in the neck. And the thought was that 80 % of the fibers there are afferent to the brain.
And that the major way station was deep, deep, deep in the medulla, the tractus solitarius, which then modulated the locus coeruleus and the dorsal raphe, these deep brain regions that were fundamental to theories of depression, certainly in terms of their role in catecholamine release, serotonin and norepinephrine and so on. But it was clearly a bottom-up approach.
Marom Bikson
So let me talk about that because I the want to put that in the context of what we talked about so far. Up to now when we talked about ECT and also TMS, electroconvulsive therapy and transcranial magnetic stimulation. First of all, we're talking about non-invasive techniques. So, these were non-involving implants.
And secondly, they are transcranial, right? That's the transcranial in magnetic stimulation, but also in ECT where in both cases, the transducers are on the scalp but we are going transcranial, which means we're trying to deliver current through the cranium into the brain. And therefore, you might think that the primary targets are our cortex. but what you're describing now is something very different where now, first of all, it's an implanted technique. It's stimulating the vagus nerve, which is this major cranial nerve, and it's targeting it as it exits sort of the cranium right here in the neck and you're stimulating it and so you're distinguishing between let's say top down, cortex down and vagus nerve stimulation which might be imagined as a sort of bottom up approach heading in this direction.
Harold Sackeim
Right, and absolutely correct in terms of characterization. even more so, one of the profound contributions of TMS to our thinking is here is this punctate, rather focal, far more focal than ECT, whose mechanism is through the generation of electrical fields in the brain, through the…in fact, induction of depolarization, but whose reach was only really, you know, a centimeter or so deep into the cortex. .So, anything that it was doing that was more widespread was a polysynaptic phenomenon. that was a consequence of the physiological changes induced by the stimulus.
But from the point of view of the stimulation technology, it was truly limited to the cortex. And my thinking about ECT always emphasize really its cortical representation largely because what we were doing when we were churning the dose up of unilateral ECT making an effective treatment was we were shifting the site of seizure initiation from the motor strip to more anterior prefrontal tissue - we were exceeding their thresholds was my view.
And so, even though certainly within the field of ECT, things like bilateral ECT, which was a gold standard, had a deeper impact, its electrical fields would involve a greater degree of the brain. That wasn't necessary to the efficacy of the treatment. was the more and more we were, because when we, for instance, reduced pulse width markedly, that was not just a manipulation of the efficiency of the pulse.
There's also a manipulation of the depth of the stimulus. People often haven't appreciated that there's a strength duration function. And as you make the stimulus weaker, either in pulse width or in current, you're going to be more focal. And so, you change the geographic distribution as well.
Marom Bikson
When you want to ask about that, that we talked in earlier part about, it the electricity that the ECT is putting in directly or is it the seizure? So now as you're talking about targeting, are you targeting the electric field or are you talking about targeting the seizure?
Harold Sackeim
Well, both because where the seizure is going to start is going to be a function of your electric field, right? And you're changing the pathways when you change the parameters of the stimulus. That's the only point I was making in this context that it was pretty superficial what we were thinking. And so, and that led to the great emphasis on prefrontal cortex, in particular, in the mechanisms of these anti-depressant treatments.
Chapter 5.2: Time Course of Brain Stimulation
Harold Sackeim
And so here was this intervention that was starting from the bottom up. And what was also very different about it was that the time course of the effect in epilepsy was something we had never seen before. It's not that people get better when you turn on VNS for the first time. It takes often months for you to see a therapeutic effect. And it turns out that that's also true in depression that we now have when you think of it, interventions for depression that display profound antidepressant effects as a matter of hours like ketamine or to some extent like ECT, or it may take months or years to see the therapeutic effect. And that was true of knowledge of VNS in epilepsy to begin with.
Chapter 5.3: Anti-Seizure is Anti-Depressant
Harold Sackeim
And so, it seemed like in many ways you were resetting a tone in the brain that had a very different path than our standard treatments. And because of my view that the anticonvulsant properties of ECT were fundamental to its efficacy, here was another new anticonvulsant, clearly VNS.
And so, the question was, was it of value? And many people, think, agree that one of the greatest discoveries of the 20th century in psychiatry was the role of anticonvulsants in bipolar disorder in particular, That what we now call mood stabilizers, the same drugs we use to treat seizures we often use to reduce the cycling in bipolar disorder, and sometimes are effective of consequence also in unipolar depression.
Marom Bikson
I just want to put a point on what you just said because it's a nice point. So, you're saying your work with ECT, the seizures produced, this then triggers an anti-seizure adaptive response by the brain and the hypothesis was that that's what was driving the mood improvement. OK Now you have another technique, VNS. Now this does not...produce a seizure, but it's by design or it's intended to activate the same anti-seizure processes. And that's where we're saying, well, therefore we have anti-seizure processes as with ECT. We now have a path towards mood adjustment.
Harold Sackeim
Right, and for what it's worth, in terms of modern development, one of the ways this has played out is we now have GABAergic antidepressant drugs that are used to treat postpartum depression. That idea that we need greater inhibitory transmission to treat the depression has become more widespread.
But that was my thinking back then. And I approached the company early on saying, let's study this in...
Marom Bikson
Then Cyberonics right? Then called Cyberonics.
Chapter 5.4 : VNS Trial for Depression, Refractory Patients
Harold Sackeim
They said they weren't interested. And I know Bob Post, who as we mentioned last time was Mark George's mentor at the NIA. He had the same idea. I mean, he was really one of the great intellectual leaders in introducing antidepressants, anticonvulsants in the treatment of bipolar disorder in particular.
And so, he was championing the idea that we should maybe consider this. And what happened was the company a couple years later went to Mark and said, are you interested in doing a study with us? And Mark said, well, yeah, I'm interested, but why don't you speak to my friend Harold? I said, don't we round in John Rush in Dallas?
And John said, let's also include Lauren Mangerelle. So those were the four sites that did the first work using VNS and mood disorders. the intention from the start...rightly or wrongly was for people who essentially had run out of options, that the population that is being considered at the beginning with almost all the brain stimulation, newer brain stimulation technologies is a resistant population.
Marom Bikson
We talked about in other parts and like you said for better or for worse, these are refractory and so very difficult to treat patients.
11:07
Chapter 5.5: Slowly Emerging Response
Harold Sackeim
Right, and VNS and DBS in particular in the mood disorders, because they're surgical, is reserved for people who have not done well with many, many things. And VNS, in part because it has some annoying side effects like voice change because of the overflow of the stimulation to the vocal cords when it's on, is also one of its limits.
But what was, we did the first study and found that about 40 to 50 % of patients after a year were showing some degree of clinical improvement, some degree of meaningful clinical improvement. So, this sort of confirmed that in depression, it's like what you did, the time scale is like epilepsy keeps.
You see further and further improvement up in those initial works up until about 18 months. Nowadays, and maybe even because it's more resistant patients, two years or longer for it to show peak clinical effects. So, this is a slowly, slowly changing phenomenon.
Marom Bikson
Can I ask about that for DBS, for VNS in general? something that takes a year to act or more, that creates a lot of burden as far as designing a clinical trial and also as far as establishing efficacy, right? Because patients are not stable. These trials not become much more expensive. There's a lot going on in these patients' lives unrelated to the trial, right? I mean, does that really raise the bar as far as the challenge of-
Harold Sackeim
Well, yeah, the bar is raised even further. I'm glad you asked that because what is also true in this population is that the likelihood that they're going to ever remit, become asymptomatic is remote. and that lesser degrees of improvement can make a meaningful difference in their lives, but for the very reasons you're mentioning, the instability of it, the life changes that are going on, all of the other stressors and so on. How do you establish that meaningful effect that becomes very, very expensive to the that largely?
Marom Bikson
So long duration adds essentially noise and patient selection makes the signal small because they're hard to treat.
13:02
Chapter 5.6: The treatment held, FDA review, and refractory patients
Harold Sackeim
And that was reflected, sort of to make the package complete. What was exciting about this technology clinically was not the number of people who benefited because it wasn't, you know, it wasn't overwhelming. It wasn't like ECT where you expect 70 % to remit here. Here we were saying maybe 40%, 50 % after a year might be responders with very few remitters, right?
And…But what was exciting about it was that if you benefited, you held it. And that was the discovery that prompted the approval by the FDA review committee, essentially the recommendation to the FDA by the scientific and medical community was to approve of this, even though the study that was presented to them, the first study was a failed study.
After our initial pilot study, Cyberonics we did, studied 60 patients at these four sites and documented that a number of them got well and that if they got well, stayed well, that sort of thing. When, when Cyberonics ended its pivotal trial, it did a 10 week sham controlled design, right? A period of 10 weeks and something that takes a year to show clinical benefit, right? With a lot of noise. So, so that was not a failed study in terms of just missing its endpoint in a design that was too short a period given the nature of the clinical benefit. But what was also manifest there is, and this was observed just the night before the presentation to the FDA committee, was that the people who did well stayed well -that everybody was followed for at least a year longer afterwards. And that was unheard of because we're writing about this now with the modern LivaNova trial.
Marom Bikson
LivaNova which is essentially the new name for Cyberonics.
Harold Sackeim
Yeah, LivaNova a very large neuromodulation heart disease company, bought Cyberonics and bought this technology. And the point is that the more treatment resistant you are, the more trials you have failed, it generally is a case that that predicts your lower likelihood of benefiting from the next treatment. But also, if you the benefit, you're likely to have relapsing.
So going back to the work in ECT, that was the work that generated a lot of our approaches to measuring things like treatment _ resistance. How do you the go about deciding that somebody's treated and what constitutes an adequate trial and not and so on. And there's now really a science of that. And it's an indication that it's a spectrum and that the people who have the greatest degrees of resistance are the ones who, for instance, in STAR-D after you failed two antidepressant drugs, prospectively, your chance of both benefiting from the third or the fourth, holding it for a year was less than 5%. So, in the current RECOVER trial, the LivaNova trial, the sham is one year.
It's a… so people are activated or not for a period of one year because of the time course of the treatment.
Marom Bikson
I want to go back to that FDA decision. So, if you can just say more about it. the benefits are durable, but is that because the device is still on? TMS, you stop your TMS, ECT. So it's good that it's durable, but it's like you're still taking the pill. Would you to talk about what happened specifically with the trial not working, but then the FDA ultimately allowing it?
16:55
Chapter 5.7 : Slow Wash-in Leads To Durable Therapy
Harold Sackeim
Sure.
Durability, why it is durable is something that we can speculate about. The major contentions are being on 24-7 means that you're going to continue to get the benefit and that there isn't poop out in the stimulation pathway, that the stimulation pathway functions in a way that it keeps the effect going. And that's very much like, I think, some of the thinking in DBS for Parkinson's, that if you stop the stimulation, you get immediate loss of benefit. But that to some extent can't be true with VNS to the extent that one, its onset is very different than the onset with Parkinson's, where in Parkinson's you see the benefit in the operating room and you decide where you're going to stimulate, which pair of electrodes you're going to turn on as a function of your immediate clinical benefit. VNS may take months to years to show its clinical benefit. we don't know, and it's certainly not the case that if you turn it off for a day or two that you're gonna see the type of relapse that you would expect with Parkinson's, so that there's an intrinsic durability, which is different.
Now, very importantly, I theorize that in depression, the longer it takes you to get well often, the longer it lasts as a general phenomenon so that the interventions, biologic interventions, other than the psychedelics, which are sort of, we could talk about a class by themselves, but if you the sleep deprivation, about 50 % of depressed people will be, an antidepressant effect within one day but because you can't keep people awake all the time, they relapse, almost everybody relapses. The same with ketamine that you can get within two hours, a marked antidepressant effect, but that antidepressant effect will last on average maybe five to seven days. It has a very short lifespan.
Accelerated TMS, now we can within five days get a profound, through Nolan's work - but it also looks like those are the people that are relapsing very quickly after obtaining that immediate effect. And that I think is also part of the issue with ECT is why you have 50 % of people relapsing and most of it is after when you discontinue the treatment and so on. So, in actuality, psychopharm and psychotherapy have better durability often than these other interventions and they also take longer. So, it's not hard to see why a neuroplastic effect, which you build over time, may have more stability than one which you have a very short life in building. that's something that we're thinking about as one of the possible issues.
And that, by the way, will be the trade-off in TMS is if this type of thinking is true is because you have better durability with standard TMS, I think, than you the with accelerated. And so you'll have to get a trade-off as to what's the optimal balance there, spreading out the treatment in time versus accelerating its improvement. Right. You can see how...
Marom Bikson
....or add maintenance.
Harold Sackeim
In any case, VNS is incredibly interesting because of, and then when you think psychedelics, psychedelics blow this all up because they're showing that single dose exposure can result in six months of, there the interpretation is that it's the nature of the reframing of one's perspective on life that is fundamental - all sorts of tests of that going on.
Chapter 5.8: Winning Over Regulators and Payers, Treatment Durability and Patient Refractoriness
Marom Biikson
So, what won the FDA over again, going back to a mixed bag, was it a split decision kind of thing?
Harold Sackeim
When the panel that heard the evidence on VNS has voted to support it. It was largely swayed by the clinical argument that…at Columbia, our patients, had averaged over were 20 drug trials. All had failed the ECT with us, patients not only showed some benefit late, but they held it. and that seemed to be convincing to them that not just that you could get movement in a snapshot in one point in time, but that it was something that was reasonably durable, which is totally unexpected in this population.
Again, to give you an example in the ongoing RECOVER trial, patients average more than 13 failed treatment trials beforehand. And their average, they're in their 18th year of their current episode, more than half of their life, 50 % of their life has been in depression. And so that's a very, very different type of patient population than, for instance, the patient populations that we think about for TMS.
And so, these are at the very other end of the spectrum.
Marom Bikson
Could you put that in context again, eventually did pass through the FDA. The fact that there's another trial being run now suggests that there was at least from the clinical community or from the perspective of the company, a need.
Harold Sackeim
Absolutely.
No, what had happened at the FDA was once this recommendation was sent to them, the FDA scientists reviewed it and rejected it. They said, you guys don't have class one evidence. We don't care how compelling it is that you have this durability. This thing does not deserve to be approved. And so, the FDA official response was to turn down the request for approval was disapproved. And that then led to a political process which played out in the halls of Congress and, in the New York Times and with Senator Grassley, who's still around, weighing in on this.
But essentially, the FDA was, I think, probably pressured as an appropriate term, pressured to rethink that and there was a meeting where the FDA administrator in charge of this had this FDA scientist who had reservations about VNS and two independent scientists, John Rush and I, who met with at the FDA on exactly this issue and hammered out an agreement that we reviewed all of the data and we reached some consensus and the FDA then decided to approve of it, to approve VNS. so it was, and within the academic community, obviously, there were eyebrows raised because of the, to get the FDA to reconsider it was a political process, not necessarily a scientific one.
And as a consequence of that…Perhaps VNS was one of the very first treatments where there, to my mind, there was a dissociation between the action of FDA and the action of the Center for Medicaid and Medicare Services, CMS. CMS had a coverage decision where it determined that it would not cover VNS for depression. This is not an inexpensive intervention where you're doing surgery, and clearly part of their worry, I presented to CMS, think on two occasions, John Russia and I both did. And part of the worry was that we use figures like 30 % of people with the major depression are going to be treatment resistant and treatment, not there's a big, again, a big avenue of treatment resistance in terms of the level that we're talking about for VNS or what would be coming to TMS. But their worry was that there would be tens of millions of people who would be considered for VNS, because of what the boundaries were, the lack of boundaries and so on. That was one of the issues. So, VNS did not receive reimbursement via Medicaid or Medicare. And when the U.S. government services the not provide reimbursement, then private insurers have no incentive to either. And so that resulted in the fact that even though it was approved, very few patients were getting it because it was non-reimbursed.
And along the way, Scott Aronson published in 2017, a highly important retrospective paper in the American Journal of Psychiatry that looked at the registry. As part of the original FDA approval, the company was required to have a VNS registry and it collected essentially over a thousand patients who had VNS and were treated with treatment as usual and patients, similar patients, highly treatment resistant patients, but who were not implanted with VNS.
And that paper reported quite meaningful differences in antidepressant effects and the longevity of those effects between those who had VNS and not VNS. And it prompted CMS to essentially ask, should we reconsider what we did and now reimburse this intervention?
And to the that as part of that reconsideration process, they actually had a meeting in Baltimore, just a methodological meeting. How do you diagnose treatment resistance? Is it reliable? Does that have predictive value? Is it something we can, and as part of their decision-making in terms of whether to reimburse it.
And this was consequential because I don't know the reality of how much this trial costs, but my guess is it's around $300 million. And with CMS funding a substantial portion of it, the surgeries, the implants and so on. And this is a study of a thousand people who half of whom get implanted, half of all of them get implanted.
Half of them get turned on, the other half don't get turned on for one year. And then everybody is followed for four more years. So, it's a five year outcomes in these highly, highly resistant patients, half of whom are unipolar, half of whom are bipolar. And it's the unipolar where we now know the outcomes.
Chapter 5.9: Lessons From Deep Brain Stimulation, "Cleaning Up the Signal"
Marom Bikson
I see. What you just described, the cost of the time, right? It goes back to what I asked before about the, when you have treatments that again, maybe take a long time to wash in and for other complicated reasons, the costs are massive. And as a result of that, it also seems like you don't get a lot of shots on, you're not going to the 10 of these a year, right? You're going to get one of these a decade. Right.
I mean, I think there's more to say with the VNS story, but I wonder if now is a good time to bring in another technique that I think emerged between the time of this first VNS trial to the one that's being run now. And that's deep brain stimulation. And I don't want to spend too much time on it, but I more wanted to ask you at a very high level, what does the experience with DBS for mood, which is an ongoing experience, The trials, what are the broad lessons from that that may apply, that are relevant both for DBS, but across neuromodulation for depression?
Harold Sackeim
Well, I would start with, think. As almost as a teaching vehicle, it's easier to start with DBS for Parkinson's and say what the broad lessons are there and what they teach us for DBS for depression or what we need to know in DBS depression. The very first thing about DBS and PD, and it may be true in depression. Is it illustrates the fact that putting electrodes in the brain and improving function may have nothing to the with what's wrong with you to begin with. And that nobody ⁓ would argue that putting electrodes in a healthy piece of tissue of the brain and buzzing it, taking offline cures Parkinson's disease.
Marom Bikson
Parkinson's isn't caused by a lack of metal electrodes in the brain, and it's not caused by a lack of extra electricity floating around in the brain.
Harold Sackeim
And even more so, we know that the primary lesion that we're dealing with is upstream to where we're stimulating. That we're not even going in the neighborhood of where the problem arises from. And we know that the problem is a result of the lack of transmission of dopamine in a particular set of synapses. And we're going down two subway stops down from that. And we're buzzing that tissue and cleaning up the signal.
Marom Bikson
....cleaning up the signal.
Harold Sackeim
Right.
So, this is a signal that ultimately gets expressed in the motor cortex and somehow by going in and one of Benabid's really brilliant maneuvers was showing essentially that, doing it at 20 or 30 Hertz didn't work. That when you went to 120 Hertz, all of a sudden people could move again. so that there's an engineering, a science to intervening in these chronic illnesses where we modify their expression that may have very little to the with their pathology. in psychiatric illness, I think it's even much more complex because life changes our circuitry in all sorts of ways. And that where things started from and why they came to be may have very little relationship to, you know, what the current wiring diagram is.
Marom Bikson
...or the paths that we can access those, right?
Harold Sackeim
Exactly.
So, and of course, ECT is a brilliant example of that because nobody ever got depressed because they lacked the seizure. and yet that may be a central ingredient in moving them. so that is one of the things, and this is one of things I think about and think about the future of brain stim because I think we can move to more fundamental interventions.
where it's no longer just shaping the expression of the illness, but thinking more about the likelihood of these epitotic things occurring and modifying essentially the substrate that allows these explosions of malfunction to occur.
Chapter 5.10: New Ways of Thinking About Mechanism, Brain Stimulation Instilling a Resistance
Marom Bikson
So related to that, - this new way of thinking - based on this new way of thinking, you might criticize the approach to TMS that views depression as, know, hypo function of some part of the brain here. And when we the TMS, we simply turn that knob up to bring that activity up and therefore reverse the sort of the disease ideology. So you would also maybe push back against, as you did in an earlier part of this interview against adopting too simplistic of an explanation again even for TMS that right there were all we're doing is reversing the dysfunction
Harold Sackeim
Right, and I think we have to be really humble. mean, one of the things I learned from the imaging work I did in ECT was it was very clear that depressed people came in with hypofunction of their frontal lobes - was tons of people at the time, Baxter and others were all demonstrating that. That was some of the most important discoveries in affective neuroscience.
So, what does ECT the? It shuts it down more. And the more it shuts it down, the more shuts it down. Right. You are. Right. Okay. _ explain that. and yes, there are ways of getting it. This is sort of almost becoming Talmudic. But,
Marom Bikson
Right...it goes the opposite way...
Harold Sackeim
It could be that what we're seeing is the reaction of the brain to being depressed, that it's trying to do what ECT does, but ECT does better. Right. There are all sorts of ways that's sort of like the hypo function of the left fixing it sort of thing, and at a much larger level.
These major psychiatric illnesses, whether they're mood disorders or schizophrenia or bipolar illness of some types, or OCD, panic attacks, they're all stress sensitive. They're all phenomena where ongoing difficulties in life can trigger.
And so, the bigger questions are, why is it that 25 % of people when exposed to trauma develop PTSD, but 75 % the not? Why does loss result in depressive states often that are unending in some, but not in others? what is it, you know, that sort of thing. And that I think is something that brain stimulation is well equipped to address in the way ultimately of instilling a resilience as a goal of treatment. there's, think, what I teach, there's wonderful preclinical animal work, which shows that we could instantly change an animal to be socially dominant or not. That is essentially the same area of the brain that we learn where we have control or not. And this is sort of going back to our discussions about what depression is. I was highly influenced by Marty Sullivan's demonstration that you can give two animals, dogs initially, the same amount of electric shock or pain, but the animal that could control it would not stop eating, would not give up on escaping and so on. But and the animal that couldn't control it went into this depressive syndrome. And that was a clear demonstration that it wasn't just being punished, but it was the idea, the abstract notion of control that she expressed even at the rodent level. So, it's obviously not propositional.
And that is fundamental, I think, to the grit and effort and determination and, how long you are in the four swim tests swimming. And that is essentially, the neuroscientists that study helplessness and so on, agreed that medial prefrontal, dorsal premedial frontal cortex was fundamental there. And so it probably is a case that the same factors that are going to help you compete in life are also going to help develop a resilience against illness. But I think that that is one of the ways that we're going to go in the future.
Marom Bikson
So, is it correct to say we're saying is instead of thinking of brain stimulation as reversing the etiology of the disease itself, so _ treating the disease, the brain has its own mechanisms for fighting these symptoms, right? And the stimulation is boosting those mechanisms, right? So, if it's resilience, then brain stimulation is essentially boosting resilience.
Harold Sackeim
Yeah. All we're doing is depositing energy, right. And we're trying to activate some circuits. So those circuits better have the type of therapeutic effects that we hope they do.
It may be, for instance, detecting whether somebody's hallucinating is a different thing than detecting what they're hallucinating about and the content of their hallucination. And we've been spending a lot of time as a field trying to figure out the content of each person's and in many ways try to undo it fundamentally, if these are expressions of amygdala hippocampal circuits, let's say, it doesn't matter necessarily what the content is or how it got there. It's at that level.
Marom Bikson
...the circuits have to be treated.
Chapter 5.11: The Birth of The Journal of Brain Stimulation
Marom Bikson
All right, this may be a very good time because we're starting to talk about brain stimulation and themes in general, maybe to transition to talk a little bit about the start of the journal, the brain stimulation journal, how that started. And of course, now the conference that's associated with it. So, I'm sure you were busy enough without having to start a journal. So what was the thinking, the motivation, the goals for the journal and how has that gone as far as what your vision was?
Harold Sackeim
Well, I think Mark and I and Josh Spieler, the three of Elsevier, were blown away by how successful this journal has been and how much impact it's had on our field and its development. It didn't take great insight to know that
Marom Bikson
Josh the publisher
Harold Sackeim
this was an emerging field of medicine that with extraordinary potential. And what was really unusual and remains unusual about our field is that it is so interdisciplinary that there's really very little distinction between neurology and psychiatry in the contributions here. This is one of the areas where it isn't or neurosurgery divided up by
Marom Bikson
and engineering and neuroscience.
Harold Sackeim
And yeah, and all of those that that all of us are in this soup. and we learn from each other and we make each other better and in a lot of ways. So that was sort of the vision at the beginning saying there isn't a place and we need to have something that will embrace
the neuroscience of stimulation. It's almost like thinking, I just saw yesterday on the ACMP's website, they have a little video from Joel Elkes, who is one of the fathers of psychopharmacology. And he was talking about what were the first meetings in psychopharmacology. And it turns out the very, very first meeting of international meeting psychopharmacology was held at the place I went to school at Magdalen College at Oxford, where Sherrington had taught and so on and you know a small group back in 1951 and that sort of was sort of a bit like the feeling of some of the early meetings on TMS and early issues with the journal to tell you the truth. I pitched the journal the idea of this journal to Josh Spieler in Toronto -I guess around 2006 or something like that. at the time, print journalism was dying because the internet was coming around and people were downloading materials. I mean, I used to subscribe to, I don't know, 10, 15 journals that had shelves and shelves in my office full of, you know, nobody that said anymore. So the business models were going away.
And on top of that, publishers were not starting new journals because they thought they were losing their pants on the ones that they had. And that the only journals they thought were successful financially for them were those that had a society. So, and then like the Society of Biological Psychiatry, which is published by their journal, Biological Psychiatry, published by Elsevier. They own the journal, that society - where and so and there are differences whether sometimes the journal is owned by the publisher or not. But my feeling at the time was I didn't think that the politics of the society was something that we needed to burden this field with to begin with. that particularly because it was so broad in its disciplines that you wanted included and represented. how would you get people to sign on to from the beginning, know, something that, right? So.
Marom Bikson
So, the _ journal that was born without an associated society. It was just a journal.
Harold Sackeim
Yeah, and no subscription built in and so on. But at the time it was, the university subscriptions were applying for it. But Elsevier went for it. at the time I was diagnosed with a pretty serious illness and felt that I couldn't take over as editor in chief.
And so, I did the really smart thing. I got a guy who did a much better job of it than I would have ever done, Mark George, invited Mark if he would be the editor-in-chief instead. I would stay on as the founding editor.
Marom Bikson
The postscript by the way, you got better. But anyways, when you got Mark George to come on
Harold Sackeim
And Mark spends, two to three hours a day, almost every day of his life devoted to the journal and an extraordinary commitment with a perspective on the field, which is, really profound and sensitive. so, he's done. And so, the journal is having to the work and seeing how your child is grown up, you know.
Marom Bikson
I see that right now.
But I want to emphasize how much Marc I mean, again, I see that now as a technology editor for brain stimulation. He reads every single submission and then that comes in with an editor-in-chief level of review, then going on to the deputy editors, then going on to editors and reviewers. So, the papers going through brain stimulation receive a level of attention, constructive feedback. You're not going to get anywhere. You know that when a paper comes out in brain stimulation, the brain stimulation is solid. You know what I mean? That aspect of it has gone just through the highest level of rigor. But go ahead. So, the journal exceeded your expectations.
Chapter 5.12: A Field That is Technology Driven / Exquisite Precision and Sprinklers
Harold Sackeim
Yeah, and that is because I think the field has exceeded everybody's expectations, the growth of this field. This is a field that...is largely technology driven. It's not that our understanding of the brain has had such a radical improvement and, you know, that we now know what to do better. It's that we can do things we could never do before. and in many ways, we are figuring out what things mean afterwards, because we're saying, we could the this and this is what happened. it must, the organization must reflect this or that.
Marom Bikson
I love that point, I want to underline you're saying the growth and the success of brain stimulation, I mean, past twenty years, especially the past decade explosion has not been driven by necessarily fundamental breakthroughs and what we know about plasticity or depression, which, we're, I'm sure there have been advancements, but probably I think the level of advancement has perhaps been disappointed, disappointing for many. What has driven the explosion and brain stimulation is the technologies that have been introduced. We can talk about all those new tech, we talked about several of those technologies so far, we have now new emerging technologies in it. So, it's those technologies, that engineering that was then put in the hands of neuroscientists and clinicians that are driving the excitement and the innovation.
Harold Sackeim
Right. Obviously, Tony Barker gave us his event. It was a tool got translated into so many different fields and so many different applications right away. it was without the tool, we wouldn't have been anywhere. And we're often limited by the tools that we have for instance, magnetic seizure therapy was a great idea.
But it's limited by the electrical engineering that we can't get a strong enough charge, so to speak, in the brain via a magnetic stimulus. It's simply. Well, no, but we could produce a seizure, but I'm saying to this day, the efficacy and the utility of the treatment are limited by its depth and degree of penetrance and so on that we have technological limits that we will ultimately be.
Marom Bikson
Until one was made.
Harold Sackeim
challenging and exceeding. yeah, the technology is, I think, is driven us. The fact that we have DBS that Benabid did the demonstration at high frequency and the other, you know, one of the very clear things about DBS is it blows people - blew me away. Is sometimes in brain simulation, we're going to need to be non-focal that…
For instance, VNS is not a focal intervention. It is impacting on nuclei that then essentially like sprinklers, sprinkled neurotransmitter across the cortex, change things like signal to noise ratios, have God knows what the longer term impact is. But in DBS, we have an intervention where if we're millimeter and a half off, we can not only not get our effect, but produce completely different effects. We're getting mania and depression when turning on the wrong electrodes. So that at some levels, there is an exquisite level of precision which is needed. And at other levels, precision may be in fact our enemy. I know that for instance, whenever I've reduced the footprint of ECT,
Marom Bikson
Right
Harold Sackeim
I've managed to make it not work. so, these are incredible tools that give us opportunities to change the brain in ways we've never could have changed before. And they are essentially what are gonna teach us about the mechanisms of therapeutics. It's inferring them from what happens as opposed to standing there and saying, I figured it out and now we need this tool to the this, right?
Chapter 5.13: New Directions in Brain Stimulation 1 - Focal Pharmacology
Marom Bikson
All right, let's finish up here by again, mean, as say, it's very hard to make predictions, especially about the future, but I want to hear your perspective about where the field is going in the next, whatever it is, the next 10 years. How much of that is tool-driven? Do we already have the next generation of tools? And now it's time to see how far we can deploy them. One aspect, we don't...
We don't even know what tools we'll be using in 10 years. But also, there seems to be old tools whose utility we haven't really, I mean, right, there's still work on ECT. There's still work on VNS. So yeah, where do we stand and how the tools play a role in this in next 10 years?
Harold Sackeim
It's one thing to think about what our targets are. Where do we want technology to go? Where do we want to invest in terms of where the largest payoff is going to be?
Non-invasive focal pharmacology will be one of the main directions of the future, as will non-invasive energy deposit. And as part of that, we'll get to is what we'll be doing to read the brain.
Marom Bikson
Can you explain first of all what you mean by the non-invasive focal pharmacology?
Harold Sackeim
Nowadays, when we take drugs, by far what we the is we administer them systemically, we swallow a pill. And typically, we have to poison the body in order to get enough of a molecule into the brain to have any behavioral effect. That's because the brain really prevents larger molecules from entering.
And so we have in psychiatry, we have the metabolic disturbances, we have the sexual disturbances, we have all sorts of things that, and it's a cross of all of, when we treat Parkinson's disease, for instance, I got so many patients from San Fon to treat with ECT who had Parkinson's because they were psychotic and they were psychotic because of the drugs that were given to treat them. And because the drugs were going not just everywhere in the body, but everywhere in the brain - and we know if you overload, with the pro dopaminergic agents, you produce psychosis and it's different system than what you need to treat the movement.
And this was illustrated beautifully. I did some work with Dan Abrams, who's the CEO, was the CEO of Cerebral Therapeutics. That company was bought by Biogen and is now the CEO of neuro insights. And Dan was a chair of neurosurgery at Children's Hospital of Denver when he lost his eyesight. He had specialized in removing tumors from the brain of children. And he humbled himself from being the chair. He did a new residency in psychiatry. While he's a resident, a fellow resident says, wouldn't it be cool if we could shoot clozapine straight into the brain?
And he said, I can the that. And what he meant by that is he had put ventricular shunts into the essentially the ventricular system of the brain to relieve the intracranial pressure thousands of times. that he could use that pathway under MRI guidance to send medication to specific area of the brain. for instance, the lateral ventricle is the floor of the lateral ventricle is the hippocampus and the amygdala. And he put a pump in the stomach wall in the abdomen with a catheter going through it into these shunts in the brain and recruited patients. He did this work in Australia. He recruited patients with epilepsy who were treatment resistant and were not surgical candidates because they had medial temporal lobe epilepsy. You don’t want to leave them aphasic.
And he had them all fail sodium valproate and he gave sodium valproate through this shunt at one tenth the dose that they failed orally. And he got 70 % seizure control, three out of 5 % seizure free.
Marom Bikson
So that's, mean, that's focal pharmacology, but that's not brain stimulation.
Harold Sackeim
That's not brain stim and it's focal pharmacology invasive. I had actually written the protocol for that company to the exactly the same thing in profound treatment resistant PTSD because of the circuitry that you can easily, and that would be a test of shutting down hyperactivity in the amygdala and the hippocampus.
Marom Bikson
And it's invasive. Correct.
Harold Sackeim
…a way to treat that, right? But you can see what the attraction is of why can't we send dopamine straight to the substantia nigra where the 80 % of cells that are dying are no longer producing it. That would be the most rational way of dealing with this illness. And I believe we're about five years away from something in that order of when human trials will begin with focal delivery.
I think right now, using brain stimulation because you can use the stimulation in two different ways. You can use it mechanically to uncage a molecule that is made inert biologically through, for instance, marrying it to a nanoparticle. You can in fact, they've now demonstrated in animals that you can make the nanoparticle, the engineering of the nanoparticle can be such that it's
Marom Bikson
…brain stimulation.
Harold Sackeim
is sensitive to different frequencies of ultrasound. so that it will aggregate like a magnet under one frequency. So, you get a hundred to a thousand times more concentration in the spot that you want. And then another frequency is used to un-cage it in that spot. this sort of thing has become routine enough, for instance, that I consulted on a study of psychedelics in Israel where we had in brain simulation, our journal, a paper indication that you could provoke the hallucinogenic experience of feeling one with the world, a sociogenic thing, by simulating specifically the insula in the context of epilepsy surgery. And there are other reasons for think that might be really a hotspot for this experience.
So now people are doing work in animals were using non-invasive pharmacological un- caging, where you send your psychedelic to this region or other regions and so on. So yes, I think that will be one way. And the other way is, this is all through ultrasound, is ultrasound has shown that it can open the blood brain barrier locally. That the lock that is preventing molecules from getting in, can be changed on a highly local basis. So that allows for a systemic administration, but only local penetration. And that is being developed particularly in the cancer area to allow neurotoxins in to particular places, but you can see it's functional.
Marom Bikson
But if we are combining noninvasive energy application to either aggregate, activate, permeabilize the blood brain bearer with a drug, is that brain stimulation? Are those papers going to be published in the pages of the Brain Stimulation Journal?
Harold Sackeim
I think it was with Mark, but when we wrote the scope for the journal back in 2006, focal pharmacology was included as part of. From the very beginning, because, you know, how much does it matter except to academics that what the packet of energy is?
Marom Bikson
you slipped it in
Harold Sackeim
I was teasing Mark recently because we don't accept things that use sensory stimuli.
Marom Bikson
…Like light flashes, auditory.
Harold Sackeim
Flash is auditory and we get a lot of that type of work that's needed. what fundamentally distinguishes flashes of light from turning on the retina on and off with the...
Chapter 5.15: Future of Brain Stimulation 2- Bioelectronic Medicine, Invasive to Noninvasive
Marom Bikson
Interesting. All right, so the future focal pharmacology. And what else? What are you excited about in addition to that?
Harold Sackeim
Well, I think that because it's so mechanical, the focal pharmacology will move quicker than the focal deposition noninvasively of energy, whether that's through temporal interference phenomena or whether it's through ultrasound phenomena. But I'll just use this one illustration to you in terms of how far our field has gone...and give you my conflict of interest.
The vagus nerves that we were just talking about, the longest nerve in the body, innervates the heart, the stomach, the spleen. And it was long thought that there were some anti-inflammatory rheumatological effects that you get from it, right? So now we have, VNS was approved just two weeks ago for SetPoint Medical, where it's cervical VNS again, where we're wrapping an electrode around the vagus in the neck. But for only one minute a day, they're stimulated one minute out of, God knows how many, and they claimed that after six months, 81 % of their patients, no longer, these are patients with rheumatoid arthritis, no longer needed biologic medications, that they had a profound anti-rheumatologic effect, sufficient enough to take people off of their drugs and persistent enough. right.
Now, and here's where the conflict of interest comes in. My son is leading a company which has been for a number of years now investigating the same idea. But where they're using non-invasive stimulation with focal ultrasound to stimulate the splenic nerve directly, the branch of the vagus that goes into the spleen with the same aim in mind. and they're trying to develop an at-home device that you use 10 minutes a day that you strap on and it detects whether you're in the right position or not and so on. And this now becomes a home-based treatment for producing the same input. And you can see that the non-invasive is huge. mean, we were talking about VNS being used in people who had failed 20 trials and most aggressive treatment and saving the treatment for that end of the illness where many people can benefit from the same thing much earlier on. And it's the invasiveness that limits our reach. So that's why I think that's important.
Marom Bikson
I think that relates to something said earlier about brain stimulation being a big tent. We talked about neurology versus psychiatry or neurosciences or engineering, but also invasive versus noninvasive. One learns from the other. One advances the other, right? Advances in one, it actually encourages advances in the other. It doesn't at all feel to me like it's a competitive, in that sense, a competitive space.
Harold Sackeim
For one thing, there really is a deep science to stimulation that applies across all of the technologies that when I'm worried about the chronaxie for neuronal depolarization and designing an ECT pulse, that chronaxie can change,
so that I think learning from each other really is a case and I think we can learn from each other also in terms of large issues.
Chapter 5.16: Combination Therapies of Brain Stimulation with Drugs, Homeostasis to Therapy
Harold Sackeim
In the therapeutics world, you have to be really good pharmacologists if you're going to be really good brain stimulation because oftentimes the people that you're treating are, are, highly complicated. They need more than one intervention in order to get well - these things are synergistic; they're often not opposed. But you also want to have a deeper understanding as to what your interventions are doing to maximize them and so on. So, some of the things I think about it are, is it naturally the case that drugs induce a alteration of biology that counteracts their potency over time? So is homeostasis likely because we get receptor down regulations and we get all sorts of other changes in sensitivity because of exposure. And in psychiatry that is reflected in poop out and in relapse and those types of phenomena. We hardly know anything about that in the neuromodulation space in terms of the systems that we're modulating. How does our modulation of them interfere with its function over time or does it sensitize it and so on. And that's a question that will be as important to TMS as to DBS and certainly any of the things that are doing with some frequency, like VNS
Marom Bikson
So, picking up on that point, the notion of combination therapies, I don't see them a lot in the literature. You usually you want to do a neuromodulation treatment, often in a way that maybe is not that translatable. You want clean patients; you're washing them off of everything. It's probably not something you the in practice. Is brain stimulation and pharmacotherapy somehow potentially complimentary and is that an area that has been, at least in clinical trials, avoided for various reasons.
Harold Sackeim
Yes, basically. If you ask questions like what meds help TMS and what meds hurt TMS, we don't have very strong answers to that, And that's the stuff that you need to know clinically. What drugs should you avoid while you're doing TMS and what drugs will benefit from it?
ECT is an easy illustration. The American psychiatric association for decades argued that we should take patients off of antidepressants during ECT because they don't help. They could only contribute to side effects, cardiac effects and so on. That was the APA's position. And the British who did those trials in the sixties where they randomized the people to placebo or drug during ECT.
They never took people off, even though those trials found no benefit from the antidepressant. So their practice never reflected their findings and the Americans practice had no findings. But we for NIMH, I chaired a multi-center study. It was published in JAMA Psychiatry, where we were trying to ask two questions.
One was the antidepressants help ECT. And two, should you start the antidepressants ahead of time anyway, because that'll give you a head start on relapse because relapse happens so quickly once you take it away. And you can't give lithium during ECT because of toxicity effects. - it could provoke a... we did a study where we randomized several hundred patients to placebo, nortriplenolone or venlafaxine during ECT. And then after ECT, everybody was treated with venlafaxine and lithium, nortriplene lithium, depending on where they and the people who originally started on placebo were randomized to _ two conditions. what And what found is the drugs helped ECT. You had a 50 % increase in remission if you were on an antidepressant.
And so that's, again, some humility for the simulation community. This is our most powerful antidepressant treatment. - and particularly people who are treatment resistant. It's not that we have all of the story or all of the efficacy that there's room for it to be shared with these modalities. And that complicates things because if you're doing science now, it's very hard in the United States to the total washouts of people because now the evidence, and there's now more evidence in the case that you're improving it when you're doing the drugs. so going almost the other way, in the RECOVER the VNS big trial, people are randomized to VNS or sham, but they also get treatment as usual, and there can be large changes in treatment as usual.
So there the science is to ask on top of everything we can do when we're trying to the our best pharmacologically, the you still see a signal, VNS? And so, we're talking about the cost of doing studies that blows it up in part because of all of the noise introduced by changing treatment.
Marom Bikson
Interesting. All right Harold… want to stop there. mean, those comments almost opened up another 100 questions. But I think we've done a wonderful job. I really wanted to thank you for spending these hours with me
Harold Sackeim
Pleasure. Are you kidding? It's fun.
Chapter 5.17: Final Thoughts, Accelerated Knowledge, BCI for Mood, and Closing The Loop
Marom Bikson
Good. Any final thoughts you wanted to leave with?
Harold Sackeim
We are really, I think, as a group, so fortunate to be at a stage where there's such accelerated knowledge because of the technologies I said this before, but, around 30 years ago, 1995, there were 200 publications a year or so on brain simulation, almost all of them ECT. And now we have ⁓ 6000 a year. And, the rate is, is extraordinary. and there's so many more. So, differences…
I guess I'd like to end with this, you know, sort of this last thought. Two years ago in Switzerland, a woman who was totally mute had a chip put in her brain and they developed a vocabulary 250,000 words for her, which they could, were over 80 % accurate in translating her thoughts into words. Yeah. And at, I think it was 60 words per minute. So this person was completely closed in on that speech. So, if we could use AI to figure out
Marom Bikson
A BCI _ yes.
Harold Sackeim
the individual words that somebody wanted to say, we should be able to figure out whether they're hallucinating. We should be able to figure out whether they're having a panic attack. We should be able to there's But more so, particularly for the episodic phenomena, we should be able to predict it just like we can predict seizures, hours ahead of time now.
Marom Bikson
Or if they're sad.
Harold Sackeim
…and we could stimulate to prevent them from happening. That's where the third area of progress that I see is not just at the level of the stimulation technology, but at the level of how we read.
Marom Bikson
And closing that loop, yes.
Harold Sackeim
And closing that loop and what is always been so problematic for psychiatry in particular in the absence of a biomarker, the determination of the illness that's being treated and its severity and its manifest and so on. so just think of what that opens up.
Marom Bikson
Beautiful. Thank you again.