Brain electrodes as depression treatment

No, we're not talking shock therapy. Researchers at the University of Toronto installed deep brain stimulating electrodes in 20 patients with severe depression who weren't helped by other means. Twelve of those individuals saw big improvement and seven experiences a "full remission." The electrodes stimulated a region of the brain called the subcallosal cingulated gyrus. Other centers around the world are running similar studies but zapping different parts of the brain.
“In the meantime we need to know why some of the patients don’t respond at all,” says (researcher Helen) Mayberg. “Are we missing the target, or are there different subtypes of the disease?” Her team is now trying to find out how to identify those who will respond to DBS, and those who won’t. “Brain surgery is not like getting your nails done, so it is important to try to find out who will benefit..."

Neurologists think that the therapy works by activating or damping down particular brain circuits. At the moment, no-one knows which of the targets within these circuits will eventually prove to be the most optimal...

The centres are also investigating the value of DBS in other psychiatric disturbances, such as obsessive compulsive disorder and addiction.
Deep brain stimulation for depression (Nature)


  1. Well, it is still kind of electroshock therapy, isn’t it? It’s just not your grandma’s electroshock therapy.

    Instead of blasting the whole head with large currents, they apply tiny current to a specific region of the brain. I guess you could say the applied current does not count as a “shock” below some threshold.

  2. Is this a, “I’ll be happy, please don’t shock me anymore” kind of happy, or a “I really don’t remember why I was sad anymore (you killed my braincells)” kind of happy. Or that, “wow, i’m so high” kind of happy?

  3. @Redmonkey: It could, y’know, be the “You turned off the part of my brain which was making me unhappy” kind of happy. I mean, the first one is right out because they were happier while being shocked, and the second one is also implausible because the shocks were presumably significantly below the threshold for damaging brain cells. As for the third, it would depend on the part of their brains being stimulated; I’d look at whether the region in question responds strongly to dopamine, and go from there. But I’m assuming the researchers already did this, and specifically chose a target that’s not merely a pleasure center.

  4. 20 patients. Twelve saw big improvement and seven experienced a “full remission.

    What happened to the other guy? Did his brain melt?

  5. This was discussed in October at a conference I attended at Emory University in Atlanta. My memory and vocabulary aren’t up to the task of describing it all, but it was fascinating.
    Basically, there are levels of depression, the lowest of which is considered beyond treatment. This treatment brings them up to a point that, if not entirely treated already, can be treated with other therapy.
    Anyway, if anyone’s interested I’ll try to find the notes I took on it and post some here.

  6. @Trimeta – I’m not sure number two is “implausible”, from the article:

    “Neurologists think that the therapy works by activating or damping down particular brain circuits. At the moment, no-one knows which of the targets within these circuits will eventually prove to be the most optimal.”

    How does electro-shock therapy “damp down brain circuits” without damaging or destroying them. I’m mean I’m not a brain surgeon, but that’s certainly what it sounds like they’re doing. Of course the article is a little generic on describing what they are doing.

  7. @Redmonkey: Neurons work by transmitting an electrical pulse down their axons. You can disrupt (or possibly cause; this is part of what’s uncertain) these pulses by sticking a metal probe near the neuron and sending some current through the probe. The magnitude of the external stimulation necessary to disrupt the neuron is lower than the threshold necessary to cause lasting damage to the neuron. After all, if these external pulses were actually killing the neurons, there’d be no question that they were damping down brain circuits.

    Of course, screwing with neurons can’t have a long-term positive effect (unless the neurons have some actual physiological defect that the stimulation is helping to regulate; we may yet find that severe depression is caused by such defects). That’s why this is currently being tested on people whose depression is so severe that all other treatments attempted have failed. The improved quality of life from deep brain stimulation very well might significantly counterbalance any potential negatives.

    @Brett Burton: I’m not sure if you were being facetious or not, but only 12 showed significant improvement; of those 12, seven went into full remission. The fate of the remaining 8 is not stated, although the quote “In the meantime we need to know why some of the patients don’t respond at all. Are we missing the target, or are there different subtypes of the disease?” suggests that those 8 showed neither improvement nor harm. I’d still like to read the paper to learn their actual results, though.

  8. Posting anonymously for a few reasons: I work at a medical OEM that is developing leads for DBS and other neuro-stimulation markets, and I reveal some personal details I’d rather not be easily collected by random web-searchers.

    DBS is fascinating stuff. The Cleveland Clinic Foundation is doing plenty of work in this area as well. (Disclosure – the medical OEM I work for has ties to the CCF)

    I have not read direct studies yet (not my area), but the “marketing” video for the treatment is flat out jaw-dropping, and very moving. In the video I saw, a patient suffering from severe dystonia went from being literally stuck in a nearly fetal position to being able to walk upright. A patient with severe tremors so bad he couldn’t even write a word on paper or walk straight went to being able to sign his name easily, draw a natural circle, and walk in a straight line.

    Depression and other mental disorders can be every bit as disabling as the physical examples above. Mine are under pharmaceutical control, and I most emphatically Do Not Want to go back to the ugly depths I plumbed before the treatment started to “work”. We don’t know near enough about the brain as we’d like, but we’re learning as much as we can as fast as we can so we can help people who really, truly *need it*. Sure, us engineers will also joke about the potential for a “drunk button” or a “happy button”, but that does not detract from the clear and present need for reliable treatment for severe Depression and other mental disorders.

    And no – pharmaceuticals prescribed for Depression, when used as directed and under the care of a physician, do not act as “happy pills”. One way I describe the effect is “putting a plank across the abyss”. It’s still there – the void stares back with at times an almost existential threat – but I no longer slump down into that abyss on the tiniest of seemingly innocent triggers.

    Sure, I’d love it if we could re-arrange society so those who have mental disorders were respected, treated fairly, and given all the support and encouragement and tolerance they need to thrive in society. I do what I can to further that idea, but in the meantime I need the pills so I can pay the bills, and companies like mine will develop DBS to help those the pills won’t help. While it would inconvenience me for quite a while (to put it mildly) if my company were put out of business because some new non-invasive non-pharmaceutical therapy or cure came along, I will be happy if that day comes!

  9. I glanced through the article and since it’s probably expensive I had the idea of a homemade alternative and applied crocodile clips to my ears which were connected to the mains supply in my house. I don’t think it’s working yet but I will try every day whilst increasing the voltage until results are achieved. I’m not depressed, never have been but it’s better to be safe than sorry.

  10. This is utterly unviable as a treatment for depression. There are perhaps 50 “deep brain” surgeons inside the United States; they are very busy individuals. There are several million people who are clinically depressed in the country.

    Frankly, there’s no time to be sticking electrodes in the brains of even the severely depressed so long as there are things like gliomas and other tumors that take precedence.

  11. >Frankly, there’s no time to be sticking electrodes in the brains of even the severely depressed so long as there are things like gliomas and other tumors that take precedence.

    There is not a linear correlation between the number of people working on a problem and the time in which that problem takes to be solved.

  12. To the several thoughtfully instructive comments above, let me add a petty one: there ain’t no such thing as a “subcallosal cingulated gyrus; that appears to be a typo carried over from the Nature piece. And as far as I know, the subcallosal cingulate is actually a subdivision of the anterior cingulate, which has been a target of DBS research for several conditions (OCD among them) for some time. I could be wrong about that though, people brains aren’t my strong suit.

  13. “How does electro-shock therapy “damp down brain circuits” without damaging or destroying them. I’m mean I’m not a brain surgeon, but that’s certainly what it sounds like they’re doing. Of course the article is a little generic on describing what they are doing.”

    The problem here is twofold i think. The technique is called electro shock therapy so it conjures up images of a system that cannot surely be natural. But current and the movements of ions across membranes inducing voltages is exactly how the brain works.

    The CNS has many mechanisms for “damping down” or “ramping up” connections without destroying them, tow examples being long term depression and long term potentiation. Connections are not of a fixed magnitude and can be altered depending on activity.

  14. I think it’s important to clarify here that it’s not 12+7 who saw some improvement but 12 who saw some improvement and 7 of those 12 who showed complete remission. Additionally, while these are patients with treatment-refractory depression, even among the “worst of the worst,” you might expect some degree of placebo response, especially to brain surgery. To see if DBS is really producing this improvement, one would need to wait (say) a month before turning the DBS device on in a pool of control patients and compare the response in that month between the two groups. I don’t know whether the Toronto group has done this, since I can’t seem to find the article, but if not, some kind of similar control would be required to determine the true magnitude of the effect. Additionally, one wonders whether the improvements disappear when the DBS device is turned off.

    While it’s somewhat off-topic, I’d also like to draw attention to an article in the Journal of Clinical Neuropsychology from 1995 that shows very high efficacy (complete remission in 4 out of 8 patients, and moderate improvement in 2 others) for the opioid receptor partial agonist buprenorphine in treatment-refractory depression. Unfortunately, while buprenorphine’s abuse potential is fairly low, for several reasons unrelated to efficacy, no further studies have been conducted. While buprenorphine is certainly not without side effects, if it’s as effective as DBS, there seem to be far fewer risks involved.

    Even more off-topic, re: the terminal man, a similar experiment has actually been done. It’s a very old study and my memory of it is somewhat foggy, but if I remember correctly, the surgeon put electrodes in pseudorandom parts of a narcoleptic patient’s brain to see if there was any area that, when stimulated, could prevent his narcoleptic fits. The button he pressed most often stimulated an area in the hippocampus, a region important for the formation and maintenance of long-term memory. Apparently, it gave him the sensation that he was about to recall something, although even with repeated presses, he never did.

  15. @Simonster: From the article:

    Advanced Neuromodulation Systems, a company based in Piano, Texas, that makes DBS electrodes, is now sponsoring a double-blind, controlled phase III trial on up to 200 patients at three centres in the United States.

    Participants in the study, called BROADEN (Brodmann Area 25 Deep Brain Neuromodulation), will have DBS devices implanted, targeting the same part of the brain as the Canadian study. Half of the devices will be switched on immediately after surgery, while the other half will wait for six months before being stimulated. Neither the patients nor the scientists and clinicians will know who is switched on at any particular time. The study is expected to take several years to complete.

    So yea, they’re aware of the problem. I’m not familiar with buprenorphine; the literature seems to concern its abilities to relieve pain and risks of respiratory depression (cessation of breathing). It’s not obvious a priori that a chemical will have fewer side effects on the brain than sticking electrodes in; the electrodes are extremely localized, while giving someone a drug might affect parts of the brain you didn’t expect. (Of course, the electrodes also require brain surgery, so it’s not obvious the other way either.) Anyway, I’m certainly in favor of more research on this subject, if there was a promising study.

    A more general note, on a topic that seems to be confusing some here: The voltages used in deep brain stimulation (DBS) are significantly lower than those used in electroconvulsive therapy (ECT), aka “electroshock therapy.” (According to the New York Times, ECT uses 70-150 volts, while DBS is more like 2-4.) With ECT, you’ve got to get through the skull; plus, you’re not targeting any particular part of the brain, so you need to zap the whole thing enough that the right parts get sufficient juice. Even then, research suggests that you’re not killing cells so much as resetting them, screwing up synapses, etc. With DBS, the probes sit right next to the neurons you want to affect, so you need much less power. Also, you don’t hit the parts of the brain associated with memory, so there should be fewer side effects. Plus, if all you know of ECT is what you saw in One Flew Over the Cuckoo’s Nest, you’re probably not qualified to judge this; in modern times, it’s only used for severe depression, the “stuck in the fetal position” cases mentioned by Anonymous #11. And it does actually work; these aren’t people who can just “decide” to be happier because it’ll keep them from being shocked again. If DBS can help these people without the side effects (loss of memory, decreased intelligence, etc.), it will be a valuable tool for psychiatrists.

  16. @Trimeta:

    Regardless of the side effects of the DBS, the risk that intracranial surgery is going to kill or mentally incapacitate someone is almost certainly orders of magnitude higher than the risk that buprenorphine will do the same. Furthermore, inserting electrodes into the cingulate is going to kill some neurons, although obviously, how many depends on the size of electrode. A year ago, I talked with a few neuroscientists who expressed some uncertainty about how much of the cognitive decline following surgical treatment for pharmacologically resistant epilepsy was due to the resection of brain matter and how much was due to the insertion of the electrodes. iEEG electrodes are likely larger than the electrodes used for DBS, but it’s still something to keep in mind.

  17. @Salvation666: On-off personal anecdotes, while tragic, are not data; this is the same sort of argument the anti-vaccination people make.

    @Simonster: According to the University of Pittsburgh, DBS is barely intracranial surgery; a 14 mm hole is made in the patient’s head, and then the 1.27 mm diameter electrode (apparently just one) is inserted into the brain, over the course of several hours. (That reference does not cite the 1.27 mm number; I found that through Google and a number of paywall-guarded journals. They also note that the actual contact at the end of the electrode is 1.5 mm in height; the rest of the electrode is shielded.) The University of Pittsburgh article also mentions some of the complications: 1% of patients suffer severe bleeding within the brain, requiring immediate invasive surgery to prevent death. 3-5% will suffer infection at the site of the surgery, requiring removal of the probe and a months-long course of antibiotics. The other side effects mentioned are comparatively minor; however, damage to neurons near the electrode is not discussed.

    Still, with one probe maybe 8 cm in total length (I’m making up this number, but it seems like the max you’d have with this sort of thing), you have a surface area of 320 mm^2 and volume of 102 mm^3; for comparison, the brain has a total volume of ~1400 cm^3 (which is 1400000 mm^3). So the fraction of total neurons which are affected is probably fairly small (less than 1/10,000 of all neurons, going by volume). Yea, you don’t want to lose any neurons, and 1% chance of “probably will die” is bad. But we don’t have the data to compare with for buprenorphine; did the study you cited discuss the side effects? And don’t forget the possibility that DBS can help individuals who are beyond all medications; even with my apparently advocacy of DBS, I’d still guess that the pharmacological approach should be tried first, unless the drug has some “definitely will happen and definitely very bad” side effects.

  18. My mother administered electroconvulsive therapy (ECT) as a psychiatrist for a while. The wikipedia page is interesting, in particular “The physical risks of ECT are similar to those of brief general anesthesia …”

    Her descriptions of the positive effects were great — people would go from literally dying because they were too depressed to move, to being able to talk with their families again. Brain injury was not involved.

    Incidentally, just to prove this isn’t your grandma’s psychiatry — another thing she did a lot, back when she was doing that kind of work, was slowly helping people come off the massive doses of anti-anxiety meds they’d been on since the 60s. Imagine having to finally cope with pain that you hadn’t been able to feel for 30 years …

  19. what are the problems of physics that prevent “virtual electrodes” from being used? If the idea is to deliver focused electromagnetic radiation to a given precise area – or wait… the idea is electron flow to stimulate biochemical ion exchange
    to propagate the impulse…. we really have to drill holes in brains and stick in copper wires?

  20. @Takuan: Actually, they’re working on that; look up “transcranial magnetic stimulation” for more details. The main problem with TMS right now is that TMS can at best affect a volume of around 500 mm^3*; while the volume of tissue affected (VTA) by DBS is apparently not well-known (or at the very least, hidden behind paywalls), one abstract I found implied that maybe it’s something like 122 mm^3. (I say “vaguely” because the reference I found is actually discussing the range of possible VTAs; in particular, “The difference in the VTA between our low (790Ω) and high (1244Ω) impedance models with typical DBS settings (−3V, 90μs, 130Hz pulse train) was 121mm^3, representing a 52% volume reduction.” You guys interpret that as you will.) Anyway, it’s a trade-off: you get much more accuracy as far as which area you stimulate when you stick in a probe, but you have to stick in a probe to do so. Of course, wearing a TMS hat would look a lot funnier than having a probe invisibly implanted in your brain, and so for the foreseeable future TMS is an outpatient procedure, not an implantable one.

    * This number is basically something I half-remember from a neuropsychiatry course I took a couple of years ago; it may be inaccurate, since I had a habit of sleeping through lectures. If someone can correct my numbers here, I’d be grateful.

  21. reminds me a little of the not-so-savory legend of the origin of acupuncture.

    How far away is mapping software so they could find the optimal external stimuli that someone induce the desired depression banishing effect? I mean in the highly scientific sense of “the brain is a squishy thing we don’t understand but we do know if you tickle here (in this guys case) with smell of onions,pinch his left pinky finger and yell in his right ear at the same time, it lights up his
    subcallosal cingulated gyrus.” In the absence of any obvious cause-effect chain I evoke bang-the-TV-set magic made possible by advances in computer modeling that now allow billions of seemingly unrelated relationship calculations. If all that gray goo is so closely wired, there’s gotta be a pathway somewhere…. OK, gotta get back to my voodoo doll…later

  22. #25 trimeta

    On-off personal anecdotes, while tragic, are not data

    Errr, while that is certainly true, pharma-funded studies about their own products are also pretty far away from reliable data.

    Like the many years of denial about severe side effects of anti-depressants “because the data didn’t support it” seem to be coming to an end, because you can only take so many tens of thousands of “on-off personal anecdotes” until your “data” becomes a farce.

  23. This is absolutely ridiculous. I sure hope they’ve exhausted every attempt to rid these people of their so-called depression before they went and tampered with their brains. I’m not talking every anti-psychotic out there either, I’m talking alternative and natural remedies and techniques. People look for the easy way out for everything, “make me a feelingless robot, anything–just please don’t force me to deal with my thoughts, feelings and real life!” I find it creepy enough being around people who take antidepressants, let alone have some shit hooked up to their brain, tweaking their neurons. That’s just f*d up.

Comments are closed.